Full text of "The Mothers (1927)"

See other formats

>S)o®> <S)o<awH)0(Si

am *$T|T srmmt sTsrm TO?ift
Bahadur Shastri Academy of Administration

MUSSOORIE

r?A

gfrWTsru

LIBRARY

| *nrrfcr

1 Accession No. _

I hpt

jj Class No.

§ Book No.

I e*e 4 od

155 ^ 646 3 _

Bf/

<>^H)oC^3> <£=£}oG^<> <£=DoG2£> *SDoG3» <^o f=^ >c$r

155.6463

100400

LBSNAA

THE MOTHERS

BY THE SAME AUTHOR

The Making of Humanity
Psyche's Lamp
The Mothers, 3 vols, 1927
Sin and Sex
Europa

Europa in Limbo
Reasons for Anger

The Decline and Fall of the British Empire
Les Troubadours et le Sentiment Romanesque

OTHER BOOKS BY GORDON RATTRAY TAYLOR
Economics for the Exasperated
Conditions of Happiness
Are Workers Human ?

Sex in History
The Angel Makers:

a study in the psychological origins of historical change 1750- 1850
Eye on Research

ROBERT BRIFFAULT

The Mothers

Abridged, with an Introduction

GORDON RATTRAY TAYLOR

Ruskin House

GEORGE ALLEN & UNWIN LTD

MUSEUM STREET LONDON

FIRST PUBLISHED IN 1927
SECOND IMPRESSION 1952
THIS ABRIDGED EDITION 1959

This book is copyright under the Berne
Convention. Apart from any fair dealing for
the purposes of private study , research
criticism or review, as permitted under the
Copyright Act 1956, no portion may be
reproduced by any process without written
permission . Enquiry should be made to the
publisher.

This abridged edition © George Allen & Unwin Ltd., 1959

PRINTED IN GREAT BRITAIN
in jo on 11 point Plantin type
BY HAZELL WATSON AND VINEY LTD
AYLESBURY AND SLOUGH

ACKNOWLEDGEMENTS

I am indebted to Mr. R. G. Phillips, who introduced me to
Robert Briffault, and to Mrs. Briffault-Hackelberg, Mrs.
Herma Briffault, and several old friends of Robert Briffault,
particularly Mme. Bradley and Mme. Stalio, all of whom
helped me with biographical information.

I should also like to pay a tribute to my two secretaries,
Mrs. Delgado and Mrs. Sawyer, who determinedly tackled
the exceptionally demanding task of transcribing and typing
several hundred thousand words of technical matter and
obscure names, as the abridgement passed through pro-
gressive drafts.

G. R. T.

CONTENTS

INTRODUCTION

page 9

BIOGRAPHICAL NOTE

The Origins of Human Society

2 7

Motherhood

The Origin of Social Feeling

The Herd and the Family amongst Animals

The Primitive Human Group

5 i

The Motherhood

The Matriarchal Phase in Civilized Societies

Division of Labour

The Institution of Marriage

Group Marriage (/)

126

Group Marriage (II)

146

Promiscuity and Individual Marriage

i 57

i 3

Primitive Jealousy and Love

173

Selecting a Husband and Acquiring a Wife

189

The Social Evolution of Monogamic Marriage

207

Tabu

234

The Totem

256

The Witch and the Priestess

269

The Lord of the Women

288

The Resurrection and the Life

306

CONTENTS

Primitive Cosmic Religion

309

The Magical Origin of Queens

354

The Great Mothers

361

Holy Matrimony

378

Modesty

386

Purity

396

Romance (/)

407

Romance (//)

423

The Mothers

431

INDEX

435

INTRODUCTION

It is just over thirty years since Robert Briffault’s prodigious work
The Mothers was first published. Its enormous length — about ii million
words when the extensive footnotes and the vast bibliography are
reckoned in — and its consequent relatively high price have prevented
more than a very few people from studying it, for few libraries hold it in
stock.

Furthermore, at the time of its first publication anthropologists were
turning away from the larger questions of cultural evolution, which
they felt to be insoluble, and Briffault’s work tended to be seen as the
last shot in a controversy which had already ceased to be interesting.

But today anthropologists are beginning to turn once more to these
larger questions. For some forty years they have concentrated upon the
detailed study of specific societies or ‘cultural configurations’ — a task
made the more urgent by the rapidity with which modem technical
progress and commercial expansion are destroying the social patterns
of non-literate and technologically backward societies. During all this
time, the attempt to erect large theories of cultural evolution has been
in disrepute. While many of the criticisms brought against the theories
of the older school of anthropologists — Tylor, McLennan and others —
were just, it is nevertheless becoming recognized that the pendulum has
swung from one extreme to the other, and groups like that headed by
Steward at the University of Illinois are beginning to develop studies of
a more synoptic character. The moment is therefore well chosen for a
reissue of The Mothers , and the time is ripe for a reassessment of
BrifFault’s remarkable work. That Briffault’s ideas might be considered
on their own merits, I suggested to the original publishers that the work
be published in an abridged form, with an introduction to put the whole
topic in perspective. They accepted this proposal, as did the owner of
the copyright, Mrs Joan Briffault Hackelberg. The present volume
therefore presents the original text reduced to about one-fifth of its
original length and omitting almost all the footnotes.

I have attempted to preserve the order, proportions, and literary
style of the original work, chiefly by omitting much of the overwhelming
mass of illustrative material and by condensing some of the more dis-
cursive comments. As Briffault gives the reader few clues as to where
he is going, preferring to let conviction grow out of the mass of the
material, I have added a few verbal signposts to assist the reader in
keeping track of the argument. Where passages are given in quotation
t.m. — i*

THE MOTHERS

marks they are in quotation marks in the original, and the source can
be found by referring to the corresponding passage in the unabridged
version.

It should perhaps be added that Briffault prepared for the Macmillan
Company in New York a one- volume work which was published in 1931
under the title of The Mothers ; it was a complete restatement of the
material and in no way an abridgement of the original English three-
volume work published under that title.

Briffault' s Thesis

The form in which Briffault couched his ideas and the relative lack of
interest with which they were greeted can only be understood in rela-
tion to the controversy of which they formed part.

This controversy was launched in 1851, when the famous jurist. Sir
A. Maine, published his Primitive Law , in which he asserted that the
patriarchal family was the original unit of society, and that larger social
units had been built up by the aggregation of these family units into
clan and tribe. In support of this view, he cited chiefly Biblical examples.
In the same year, the Swiss jurist Bachofen was preparing his Das
Mutterrecht , asserting that the original state of man had been one of
sexual promiscuity, from which had emerged matriarchies, which had
only later been replaced by or converted into patriarchies.

This set off a series of attempts to draw up schemes designed to
account for the whole development of human society, and represented
the application of the idea of evolution, which had proved so fertile in
the biological field, to society as a whole. J. F. McLennan made the most
important restatement of the matriarchal view in 1886, citing a great
mass of new anthropological evidence. Early in the nineteenth century,
Westermarck — a man without anthropological qualifications — published
his History of Human Marriage , in which he attempted to re-establish
Maine’s position. He was not so much concerned to draw a picture of
the whole development of human society as to assert that lifelong
monogamy was the normal pattern of marriage throughout human
society, polygamy representing a degeneration from the original mono-
gamic pattern. This thesis was naturally much to the taste of Christian
apologists and traditional moralists generally. Westermarck’s works
enjoyed wide acclaim, and he wrote a long series of works embroidering
this theme, most of which are still to be found in the majority of public
libraries. Largely as a result, this view of marriage is still held by very
many laymen, in so far as they concern themselves with the topic at all,
and is often given fresh currency by American anthropologists. It is
asserted in the new Chambers’s Encyclopaedia.

This view is, in point of fact, wholly untenable, and there can be

INTRODUCTION

little doubt that Briffault felt incensed by Westermarck’s scientifically
unjustified success, and that his main object in writing The Mothers was
to explode this fallacy. This he undoubtedly achieves, adducing such a
wealth of material to the contrary, and so decisively convicting Wester-
marck of manipulating his references and betraying other signs of bias,
that one might have supposed that belief in the universality of mono-
gamic marriage would have been abandoned for ever in a gale of
laughter. In fact, as we know, his statement was ignored; and if this new
version of his work does something to restore a more detached and
speculative approach to the topic, it will have been worth while for this
reason alone.

But Briffault was not content simply to destroy — he sought to estab-
lish an alternative theory. In contradiction to Maine, he asserted the
former existence of a primitive matriarchy universally preceding patri-
archy, but, unlike Bachofen, he did not define matriarchy in terms of
actual mother-rule or inheritance through the maternal line, but in more
general terms as a period in which women were socially predominant.
As the crucial factor, he selects, for reasons which he adduces at length,
the question whether, after marriage, the wife resides at her husband’s
abode or the husband at the wife’s — i.e. what anthropologists term
patrilocy or matrilocy. (The distinction is important, and several sub-
sequent investigations designed to prove or disprove the matriarchal
theory have gone astray through ignoring it. 1 )

Furthermore, his inquiries lead him to the view that marriage was
originally a contract between groups, in which it was agreed that a man
of one group might have sexual access to all the women of another
group while being denied access to his own. This primitive restriction
on sexual behaviour was then elaborated into such forms as restriction
to all the women of a particular family — that is, in our terms, if a man
marries a woman he is thereby married to all her sisters. Briffault
persuasively argues that ‘sororal polygyny’ and its complement ‘fra-
ternal polyandry’ are not perversions of the basic idea of a monogamous
marriage. Quite to the contrary, they are restrictions of an original group
contract so different from marriage as we know it that we shrink from
applying the word marriage to it. Indeed, where such group contracts
are found they are usually termed by the scandalised Western observer
‘sexual communism’ or ‘promiscuity’.

To establish these two main points, Briffault seeks to show that the
alleged change from matriarchy to patriarchy was associated with the
change from hunting to agricultural production and the essential
emergence of property.

Into this landscape he also attempts to fit such well-known anthropo-
1 E.g. R. Karsten: Origins of Religion (1935).

THE MOTHERS

logical puzzles as initiation ceremonies and the prevalence of lunar
symbolism.

Briffault develops his work in four main sections. In Chapters I to 5
he considers the domestic arrangements of animals, from which human
arrangements presumably developed, and seeks to show that they were
matrilocal in character. More precariously, he argues that the male
instinct created the group or herd, while the female instinct created the
family. Since he is going on, later in the book, to argue that the family
is a feature of patriarchies, the relevance of this section is obscure, to
say the least.

In Chapters 6 to 16 he considers the whole question of the emergence
of marriage, arguing the universal existence of a primitive matriarchy in
the prehistory of all peoples, and seeking to show in various ways that
marriage was originally matrilocal.

In Chapters 17 to 24 he considers various indirect forms of evidence.
In particular, he argues that lunar deities are indications of a primitive
matriarchy, and that such cults were originally served only by women,
who were the first hierophants. He seeks also to fit the concepts of
totemism and taboo into his system, and here he is least successful,
though he makes a number of interesting points.

Finally, in Chapters 25 to 30 he attempts to trace the growth of the
modem Western conception of marriage as a sacrament, as a cultural
remnant of the idea of a holy marriage between a deity and a woman.
Ideas of this sort are indeed present in our thinking to a much greater
extent than most of us realize, and Briffault’s demonstration is fascinat-
ing; but on the sources of this need to preserve a sense of sacredness —
so notably absent in many other fields — he has nothing to say. In con-
junction with this, he traces the origins of the notion of romantic love.

Criticism of Briffault' s Views

For some time before Briffault reached the point of publication of these
views, anthropologists had been becoming increasingly critical of such
attempts to provide a general schematic account of the development of
society, on two general grounds.

Firstly, as anthropological studies developed it became even clearer
that one could only hope to understand the meaning of a culture item
by considering it in relation to the whole culture. A given social action
may carry quite a different connotation in one culture from that which
it has when part of another. Consequently, to pick out a given pattern —
let us say, mother-in-law avoidance — from a number of different cul-
tures and compare them was an unacceptable technique. (Like many
before him, Briffault relies heavily on this ‘comparative method.’)
Anthropologists therefore rejected the whole method, and turned to

INTRODUCTION 13

studying individual cultures in detail, seeking to profit from this new
insight into their structure.

This argument was reinforced by another: we cannot assume that
societies which are technologically primitive resemble equally primitive
societies as they existed thousands of years ago. A long sequence of
social changes may have occurred — the marriage customs of the techno-
logically primitive Australian aboriginals are so complex that it seems
certain they represent the outcome of a long process of elaboration.
Hence (they felt) we have no information about primitive society,
except what archaeology may uncover, and speculations about the past
state of man are vain.

No doubt there is much truth in both these views. The horse has
changed greatly from the Eohippos from which it is descended. But it
is unnecessarily defeatist to say that nothing can be learned from such
studies. If we had no skeletons of the Eohippos to go upon, we might
not appreciate that the horse was so much larger, and we could only
guess that the Eohippos had not developed the horse’s specialized hoof;
but we should at least be sure that the Eohippos did not swim or walk
upright.

The passage of time may modify or complicate a given social pattern;
it is not proved that it can bring about a change in kind. Indeed, it
seems to me that it may be the case that technological advance is only
possible when certain underlying changes of attitude or psychological
make-up occur; such changes would also effect changes in social patterns
generally, so that some aspects of social pattern may be correlated with
the level of technical development, after all.

Though anthropologists tended to reject these synoptic attempts on
a priori grounds, subsequently the advance of archaeology went far
towards confirming their scepticism. It was observed that agricultural
peoples, driven by population pressures out of Asia Minor into the
steppe country, became pastoral — which is just the contrary of what
Briffault asserts to be the normal process. Gordon Childe subsequently
showed that various primitive peoples have passed through matriarchal
and patriarchal stages in varying orders, and have adopted various
methods of subsistence in quite a haphazard way . 1 Meanwhile anthro-
pologists have noted tribes which, at the present time, are passing from
patriarchal to matriarchal patterns of society; for instance, the neigh-
bours of the Tsimshian in North-west America . 2

It is now past all reasonable doubt that society does not evolve
according to one single standard line of development. Of course, it
might still be true that there was a normal line of development from

1 V. Gordon Childe: Social Evolution (1951).
a J. R. Swanton: Social Organisation of American Tribes (1905).

THE MOTHERS

which societies would occasionally depart in exceptional circumstances.
The foregoing evidence does not justify the conclusion that all attempts
to find any system in such data is vain. But once we admit that matri-
archy can follow patriarchy, much of Briffault’s material becomes
ambiguous. When he draws attention to signs of an earlier matriarchy
in a society which is now patriarchal, he may in reality be observing the
signs of a future matriarchy which is only just developing.

A New Assessment

Briffault did himself much disservice by claiming too much : it was in
the nature of the man to prefer the sweeping generalization, and he
loved to shock the unimaginative out of their preconceptions. His data
do not justify him in making the assertion that matriarchy always and
everywhere preceded patriarchy, even if wc neglect the facts just
adduced. Even if in existing patriarchies signs of earlier matriarchy can
be detected, this does not prove that a still earlier patriarchy may not
have preceded the matriarchal phase. The periods of time in question
are but a few hundred years — the pre-history of man runs to tens of
thousands.

Briffault would have made a more convincing contribution if he had
confined himself to asserting that in every patriarchy the existence of a
previous matriarchal state can be shown or inferred, leaving unbroached
the question of what had preceded that state during the thousands of
years of pre-history. For the realization that these social patterns are
labile is a novel and important one, to which we are only now coming.
It never occurred to Maine that the Jewish patriarchy which he so much
admired, and thought was fundamental and God-given, had actually
developed out of an earlier mother-centred system, as Briffault shows.
Even today, few people are any better informed than Maine. The belief
that patriarchies were always patriarchal is almost universal. Briffault
might justifiably have written his book to prove this one fact.

Again, Briffault invited ridicule or neglect by grossly over-generaliz-
ing his theory of marriage. It is certainly true that monogamy is not the
universally preferred pattern; but it is going much too far to assert
group marriage as universal. The likelihood is that humanity found a
number of different solutions to the problem of regulating the relations
between the sexes; group marriage may well have been one, perhaps
even the most widespread one. The hypothesis certainly enables Briffault
to reduce a great deal of otherwise baffling material to coherence, even
if it does not explain quite as much as he claims. It is therefore well
worth much closer examination than it has so far received.

His larger theory, in which the change from matriarchy to patriarchy
is linked with changes in the method of subsistence, from pastoral to

INTRODUCTION

corn-growing and so forth, again attempts too much in asserting a single
sequence of development. The more modest task of exploring whether
certain social structures are always associated with certain modes of
subsistence would have been more rewarding.

The Psychoanalytic Clue

Briffault’s greatest mistake, one cannot help feeling, was to dismiss as
valueless the entire contribution of Freud, for it is precisely Freud who
could have helped him to solve the points on which he stumbles most
hopelessly. First and foremost, Freud provides a comprehensive and
consistent theory of the origin of incest fears. Since, as Briffault accur-
ately notes, the whole system of exogamy rules is simply a system of
incest-regulations, it is strongly supportive of his views that Freud
attributes this preoccupation with incest to a preoccupation with the
mother. That is, Briffault’s contention that exogamy rules arose in
societies in which mothers were dominant is completely in harmony
with Freudian theory. Conversely, the jealousy which Westermarck
thought a universal human instinct is revealed by Freudian theory to
spring from a preoccupation with the father, and thus to be character-
istic of patriarchal but not of matriarchal societies. Briffault, who justly
rejects Westermarck’s view on this point, could have proceeded to
explain just why it is found in patriarchal societies, and indeed why
Westermarck should have held such a view, had he not been so cavalier
towards psychoanalysis. Unfortunately, Briffault derived his psychology
from the teachings of Shand, now almost forgotten.

Again, Freud provided, in his description of the mechanism of pro-
jection , a theory which explains perfectly why those who are preoccupied
with the mother-figure tend to envisage their deity as a mother, while
those preoccupied with the father tend to postulate father deities.
Briffault, who traces the tendency for the moon to be seen sometimes as
a male and sometimes as a female deity (and occasionally as both con-
jointly) could — had he realized this — have related this dynamically to
the corresponding social changes from matriarchy to patriarchy. Simi-
larly, Freud’s account of decomposition — the process by which people
sometimes classify people into good and bad figures, and have difficulty
in seeing that good and bad aspects can be combined in a single person
— is accurately reflected in the way in which some peoples divide their
deities into good and bad, God and Devil, while others feel that a deity
may have good and bad features simultaneously. Briffault notes the
anthropological fact quite correctly, but makes heavy weather of fitting
it into his system because he does not understand (what Freud could
have explained to him) the origins of these alternative attitudes.

Briffault was correct in his insight that the description of the deity

l6 THE MOTHERS

could be used as a clue to the social structure of the people making the
description, but tries to link the two in a mechanical way, instead of
regarding both as reflecting an unconscious attitude. (In matriarchies,
women have the priestly role, and the moon is their patron because it
seems to control their menstrual periods, his argument runs.)

Freud’s attempt to account for totemism and to explain taboos seems
to me only partly successful . 1 On the other hand, Bettelheim’s analysis
of Australian initiation ceremonies, and the comparisons he makes with
the disturbed children he has studied in the USA, seem to me the only
writings which make any sense on this subject and, indeed, to be
stimulating in the extreme. To say, as the anthropologist generally does,
that initiation ceremonies mark the transition from youth to manhood
may be true — if we assume that this transition takes place at puberty —
but it certainly does not account for the frenzied violence which often
accompanies them. It certainly offers no clue as to why the Australian
aborigines should make a long and deep gash on the underside of the
penes of these boys, and dress them in women’s clothes — hardly a
gesture designed to bring out their manhood. On the contrary, as
Bettelheim points out, it is clearly a ceremony designed to turn them
into substitute women, and it is the women who insist on this ceremony . 2
The anthropologist who is unwilling to accept a psychoanalytic inter-
pretation of such odd actions would be more honest if he were simply
to admit that he has no explanation to offer, instead of talking earnestly
about rites de passage .

Briffault, though he takes a different view, is no more a propos. He
argues that these ceremonies are designed primarily to demonstrate the
young man’s fitness to support a wife. This explanation at least recog-
nizes the vital fact that it is the women who insist on these ceremonies.
But tortures which maim, render impotent or even kill the victim do not
really have this effect, and are clearly in a different category from de-
manding (as may also happen) that the young man demonstrate his
prowess at hunting or fighting. At best, the aspect which Briffault
stresses is but a single one ; the other features — all the stranger because
they seem to have no utilitarian value or even to be harmful — demand
explanation.

Briffault is in even deeper water when he tries to explain the changes
in the status of women. It is when men come to possess so much wealth
that they can keep women in idleness that they become sexual play-
things and lose status, he declares. He realizes that this view is quite
inconsistent with the depressed status of women among the Australian
aborigines, and suggests that this is because the aborigine has used his

1 S. Freud: Totem and Taboo (1919).

a B. Bettelheim: Symbolic Wounds (1955).

INTRODUCTION

superior strength to dominate his women. But elsewhere Briffault has
argued that women are not only stronger but also fiercer and more
cunning than men. And even if this were not so, it would still leave him
under the obligation of explaining why, in matriarchal agricultural
societies, men do not equally exert their strength. The Celts, too, whom
he sees as matriarchal and deferring to women, had notable heroes;
why did they tolerate their women’s arrogance and sexual freedom ?
Finally, in our own day, in the West, man is more than ever able to
earn sufficient to support his woman in idleness; but the American
woman, for one, is hardly dominated by her male.

But here psychoanalytic theory provides a scheme which, though
derived from quite other data, fits the anthropological facts as if it had
been made for the purpose. The Oedipal situation, as described by
Freud, accounts for men’s fear that women will betray them sexually,
and their sense that they are a threat to be kept under control. But the
Oedipus situation can only exist where a strong father-figure is present,
and is intensified if he is severe or thought to be so. Hence we should
expect to find this attitude to women strongly marked in families of the
patriarchal type, and absent in those families where the children are
brought up by the mother alone (usually with the help of her brother),
and where the biological father performs no parental role. It is many
years now since Malinowski reported just this absence of sexual guilt
and concomitant freedom of women from the Trobriand Islands, where
the family structure is of the type just mentioned.

(Note how, in the West, the status of women has risen and sexual
freedom has increased in proportion as the patriarchal nature of the
family has declined.)

In making this estimate of Briffault’s work, I am naturally influenced
by my own speculations on these matters put forward initially in 1949,
and developed in 1953 an d I 95^; 1 in them society is postulated as oscil-
lating irregularly between phases in which the mother-figure is dominant
and others in which the father is dominant, with the possibility of a
balance between the two. Institutions, such as marriage or the inheri-
tance laws, change so slowly, that institutions appropriate to a father-
centred phase may persist into a mother-centred one, and no doubt the
reverse also occurs. Hence we cannot safely classify a society by the
little we know of its institutions, such as, whether marriage is patrilocal
or descent patrilineal. Thus, as Margaret Mead has shown, the Tcham-
buli have all the social features — such as patrilineal descent — associated
with a patriarchy, but in fact the women dominate the men. 2

1 G. Rattray Taylor: Conditions of Happiness (1949); Sex in History (1953);
The Angel Makers (1958).

* M. Mead: Sex and Temperament in Three Primitive Societies (1935).

l8 THE MOTHERS

For the same reason, when Briffault succeeds in showing signs of the
existence of a preceding matriarchal phase in a patriarchal society, this
does not seem conclusive proof of a primitive matriarchy, since an even
more primitive patriarchy may have preceded it.

Does this mean, then, that we should reject Briffault’s contribution as
worthless? That is the conclusion which some contemporary critics
reached; but in my view this is to throw out the baby with the bath-
water.

The Problem Restated

The task which has fascinated so many anthropologists in the last
hundred years — the attempt to devise a comprehensive account of the
sequences of social development — turns out to be insoluble and perhaps
meaningless. But this does not mean that this whole area of inquiry
must now be abandoned ; it means only that the task must be reformu-
lated. The many extraordinary social phenomena which Briffault
chronicles remain for the most part without any satisfactory explanation.
Is it a matter of pure chance whether a given tribe adopts polygamy or
monogamy, exogamy or endogamy? Or are such practices related in
some way to its mode of life, or to its religion ? Is it a matter of pure
chance whether it adopts a father-religion, a mother-religion or a
religion of some other type ? What is the explanation of the savagery
which so often attends initiation ceremonies ? The remarkable range of
human behaviour which Briffault records in such profusion still calls
aloud for clarification, and if we object to Briffault’s synthesis then we
admit the need to find some alternative.

More than this : is it not possible that the explanations of many such
cultural features are tied together ? To put it differently, is it not pos-
sible that there is only a limited number of basic sociocultural patterns,
that the almost infinite variety of those we know consists only in varia-
tions on a few simple themes ?

If such patterns could be found, the task of analysing the history of
cultural evolution would be greatly simplified. And if the conditions
determining which pattern would be adopted could be established, it
might be possible to make more reliable inferences about the social pat-
terns of human communities in prehistoric times. Only by exploring the
subject in this kind of way do we seem to have any hope of forming ideas
about the manner in which social and sexual institutions have emerged
in the long evolutionary scale of man’s slow assumption of humanity.

If the psychoanalytic approach is adopted, we require, in order to
account for the social changes which we observe on the historical scale,
only to account for the changes in family structure. Perhaps it could be
shown that economic factors make it inevitable that a pastoral society

INTRODUCTION

should be patriarchal. Perhaps, however, it may be the case that a
patriarchally-minded individual prefers to occupy himself with flocks
rather than with agriculture. We are most likely to find the answer to
such choices when we can locate cases in which a change is actually
occurring or is known to have occurred. Why are the neighbours of the
Tshimshian, for instance, moving from matriarchy to patriarchy ?

For the past half century, anthropology has taken a static rather than
a dynamic approach : it has explored the structure of given societies in
great detail. It has given little attention to the interaction between
societies and to the change of societies with time. This is natural enough,
and probably the harder task could only be fruitfully attempted when
the easier had been performed. Naturally, too, the psychoanalytic tool
has first been employed in order to explore the structure of given
societies in more detail (e.g. by Kardiner, Linton, du Bois and others).
But the time has come when it could be turned on these larger problems
of historical change.

It is reasonable to believe that a new epoch in anthropology is now
opening up. When we really understand the dynamic connections be-
tween social institutions, such as marriage, and economic conditions,
and the connection of both with personality structure, we may reach a
position from which we can attempt the task prematurely undertaken
by Maine, McLennan, Brifiault and others, of inferring how first social
institutions were developed by the human race, and what form they
may conceivably have taken.

Conclusion

Brifiault opens the way for some development on these lines both by
the devastating way in which he clears the ground of the forest of
misconceptions which have grown up and enables us to view the data
with less ethnocentrically-prejudiced eyes; and also by the many odd
features of the terrain which he then points out, and the stimulating
suggestions he offers to account for them.

It must be conceded that he is open to criticism in matters of detail.
He not infrequently contradicts himself, and sometimes uses a fact to
prove one thing at one stage and to support an equally plausible but
quite different view at a later point. He is sometimes guilty of selecting
his references to prove his point and glossing over those which are
incompatible with it. The captious critic could make him look small,
and it is easy to be persuaded that the writer who is inaccurate in detail
is, ipso facto , wrong as regards his thesis. In point of fact, the kind of
mind which is capable of conceiving a large theory is apt to be impatient
of details ; the mind that concentrates on details generally fails to see the
wood for the trees.

THE MOTHERS

We do not read Briffault for a text-book statement of incontrovertible
fact, but for a challenging argument supported by a mass of fascinating
detail. We can hardly expect that, having written a million and a half
words, having impoverished himself and damaged his health in the
process, he should then sit down and spend the next ten years trying to
disprove his own theory. It is for others to raise the objections, and to
see whether they can be met by minor modification of the main hypo-
thesis or not.

Nor do we read Briffault exclusively for his main thesis; his inquisi-
tive and radical mind explores many byways of anthropology, always
throwing light, challenging preconceptions, and offering new insights.
He is as instructive when he is discussing the origin of human clothing
as he is when evaluating the role of the troubadours. Not the least
fascinating feature of The Mothers is the way in which he weaves
material from Biblical Jewish history or from Classical Roman and
Greek sources into the general anthropological picture. It is intensely
stimulating to see societies which we have come to know and take for
granted in our schooldays, and which have thus acquired a special status
in our minds, compared with societies which we regard as strange or
primitive objects of anthropological study, and to see how they, too,
are just as strange, their development just as complex. Just as our read-
ing of the Bible, coloured as it is by religious presuppositions of a later
date, seldom reveals to us the moon and mother-worshipping origins of
Jewry, so our reading of the Classics is focused on a late phase of the
society, and its curious matriarchal origins escape our eye.

BIOGRAPHICAL NOTE

Robert Stephen Briffault’s life starts with a mystery. He seems to
have been bom in Nice in 1873, although he later claimed to have
been born in London in 1876. Briffault’s mother — we should start
by considering his mother, surely — was a Scotswoman of strong
religious views; the daughter of a sea-captain. Nothing else is known
of her life before she married — astonishingly — an elderly French
diplomat, already retired, Charles Frederic Briffault. It was his second
marriage, and of his first nothing is known either. Charles F. Briffault
had been Chef du Secretariat du President de la Republique Napoleon
in 1849, but left the service of Louis Napoleon on his unconstitutional
seizure of power, and became naturalized British. He travelled a great
deal on the Continent, where he had many friends. For many years he
made his home in Florence, and it was there that Robert Briffault
attended a private school for English boys until his ’teens, though he
completed his education in England.

At the age of 19, his father being dead, Briffault went out with his
mother to New Zealand, where, at 24, he married one Anna Clarke; by
this marriage he had one son and two daughters. After much indecision
as to a career, he decided to study medicine, and in 1901 he received an
mb at Dunedin University, and a ch.B at Christchurch in 1905. He
then set up in practice. After a visit to Europe, he returned to New
Zealand, and was president for a time of the Auckland branch of the
New Zealand Institute.

On the outbreak of the First World War, he returned to England to
enlist, and served at Gallipoli, and later in Flanders and France, where
he was rather badly gassed.

He had written a good deal in his spare time, without publishing
anything, and while in the trenches he worked on a survey of human
evolution which was published in 1919 as The Making of Humanity ;
Psyche's Lamp , an outline of psychology, followed in 1920. His wife
having died at the end of the war, he established himself in London,
where he alternated between frenzied spells of work on The Mothers ,
and working as locum tenens at one of the London hospitals. The death
of his mother in 1924, followed by the death of one of his daughters,
with other mental stresses, affected him so that at times he would slide
to the floor in a trance lasting ten or twenty minutes. After sustained
efforts, in which he gave up all medical work and depended largely on
his daughter. The Mothers was finished and revised, appearing in 1927,

THE MOTHERS

when he was 54. With this he gained some reputation. Other books
followed.

In 1930 he married Herma Hoyt, an American some twenty-five years
his junior.

In 1931 he went to the u.s., but by the end of 1932 was living in Paris,
where he remained until the end of the Second World War. While in
New York, friends suggested that he should write a novel; the result
was Europe which, to his surprise, became a best-seller in several
languages. A sequel, Europa in Limbo> followed, with other novels, and
a number of essays.

In Paris, during the Occupation, Briffault was twice imprisoned for
short periods. His principal difficulty was that he was cut off from the
English language market, and presumably from the receipt of royalties
also, and had to support himself with translations and hackwork. He
also wrote, in French, Les Troubadours et le Sentiment Romanesque ,
which may be regarded as an expansion of Chapters 28 and 29 of The
Mothers. After the war, he returned to England, in broken health,
although he travelled restlessly several times to France and Italy. His
wife, herself ill, returned to New York, believing that the State Depart-
ment would then grant her husband an entry permit — this having been
refused on earlier applications ; but it remained obdurate. In December
1948 Briffault died of pulmonary tuberculosis at St Helen’s Hospital,
Hastings, aged 75.

Briffault’s tastes were those of the Romantic. His favourite composer
was Wagner; his favourite painter Hieronymos Bosch; his favourite
play Goethe’s Faust. Poetry he loved. His intelligence was encyclo-
paedic. He spoke more than seven languages — French, German, Italian
and English fluently. In addition he knew Latin, Greek, Dutch and
Proven9al French. He started learning Russian at 70.

As a general practitioner, he was much loved and even adored by his
patients, particularly women, and by the nurses with whom he came in
contact. The medical profession looked at him somewhat askance, it is
said, because of his radical ideas and sometimes unorthodox methods of
treatment. He seems to have been reasonably popular with fellow
officers and men during the war.

Of Briffault’s childhood and the forces which may have influenced
him into undertaking the series of broad studies which culminated in
The Mothers , we have no direct knowledge. Europa tells the story of
Julian, the son of a retired diplomat living in Rome and his puritanical
Irish wife — and from it we may gather some clues. The incident in
Europa in which Nietzsche puts his hand on the head of Julian saying
‘Thou mightest be He’ is drawn from Briffault’s own youth.

Julian is, as a boy, vaguely uneasy in the cultured setting of his

BIOGRAPHICAL NOTE

father’s many distinguished, wealthy friends. He is sensitive to the
contrast between the ease of his own surroundings and the poverty of
the peasants. He is much struck by Darwin’s theories, and feels that
knowledge is the key to the mystery of life; he must know everything.
He asks a sympathetic teacher if it would be any good his reading the
Encyclopaedia Britannica from cover to cover. His teacher tells him that
nine-tenths of what is taught is untrue, and he must find out for himself
what is true and what is not. A Catholic priest, attending his parents,
finds him reading The Origin of Species and snatches it from him in a
fury. A cardinal is brought in to dissuade him from accepting such ideas
in a more subtle and intelligent manner, and he is shortly sent off to
school in England. Here the science master lets him see the glories of
the stars through his telescope ; he is awed by the insignificance of the
earth and the individual. This master warns him that the school does not
teach what are now known to be historical facts, such as that Aeneas,
Homer and King Arthur are not historical figures, for if the authority
of tradition were flouted, all authority might be upset, even belief in the
historicity of the Bible.

We can see here the gradual unfolding of doubt, and the desire to
establish a scientifically tenable world view, which evidently animated
Briffault’s first books, although the Julian of the novel does not proceed
to write works of this kind. If we ask where this curiosity came from,
we may find the answer in the account Europa gives of Julian’s aware-
ness of sexuality. As a child he is represented as suffering considerable
shock when a peasant woman, seeing him relieving himself against a
wall, jokingly asks him if he knows ‘what that is for’, with other crude
innuendoes. Sexuality breathes in the warm Italian air, and the adoles-
cent Julian’s sexual education includes a visit to a private collection of
phallic emblems. While a psychoanalyst would certainly look for some
earlier experience as the real starting-point of sexual shock — doubtless
something concerned with the relationship between his parents — we
can, in any case, see here the association between sexual curiosity and
scientific curiosity which is asserted by Freudian doctrine. This, then,
explains the possibly excessive interest in sexual matters which runs
through The Mothers , and the violence of his attack on sexual ignorance
in Sin and Sex and other writings. We can see, too, why the Church,
and especially the Catholic Church, seemed to him to be the obscuran-
tist force which opposed both scientific knowledge and sexual under-
standing.

Europa also depicts the disgust which Julian experienced when he was
transferred from the cultured, adult world of Rome to the narrow and
barbarous surroundings of an English school where, though not un-
popular, he was an oddity who knew more Latin than the master

THE MOTHERS

2 4

teaching it, but nothing of cricket and who had never heard of King
Solomon's Mines. This experience may well have reinforced the resent-
ment which Briffault was developing for all established authority and
morality, and against Britain, the country to which he nominally
belonged.

This resentment was voiced in his book The Decline and Fall of the
British Empire (though it must be noted that the title was devised by his
American publisher, over his protests). A close friend has told me that
his attitude to Britain was in reality highly ambivalent; at bottom he
loved Britain, and he returned to his country to die. It was because he
felt that Britain had rejected him that he hated it; such feeling must
have been intensified by the grudging recognition accorded to The
Mothers .

A psychoanalyst would surely see in this the reflection of an attitude
towards parental figures, since native countries are either ‘mother
countries’ or ‘fatherlands’. And it is not unreasonable, when a man has
devoted seven years of strenuous work to arguing the importance of
mothers as against fathers, to ask whether he betrayed any marked
attitudes towards his own parents.

Mrs Herma Briffault, however, informs me that he had little apparent
attachment to his mother, but often spoke of his father. (He seems,
nevertheless, to have had a picture of his mother in his room in the last
decade of his life.) His mother was evidently a reserved, ‘canny’ Scots-
woman, of strict views, whose capacity for personal warmth seems to
have been limited, and perhaps he felt, as an infant, denied the accep-
tance he desired. He certainly rejected violently his mother’s strong
religious beliefs and teaching. Clearly there was strong ambivalence here,
and the attacks — which suggest narcoleptic stupors — which assailed
him at the time of his mother’s last illness seem consistent with the idea
of a powerful love-hate relationship. His attitude to his mother country
was, as I have related, very similar, and the fact that he was brought up in
isolation from that country, and regarded as an oddity when he visited
it, may have completed the parallel.

His father seems to have been a broad-minded and cultured man; he
wrote (probably at Louis Napoleon’s request) a novel, The Prisoner of
Ham (1846), in which may be found ideas and principles which re-
appear in the work of his son — such as faith in the power of truth, and
dislike of those who proceed in everything merely from day to day,
unable to conceive any grand idea or to create anything grand or useful,
and convinced that poverty is unavoidable and evil necessary.

Briffault was generally regarded as having Communist views, and this
is no doubt why he was refused entry to the u.s. He was certainly
extremely interested in ‘the Communist experiment’, and learned Rus-

BIOGRAPHICAL NOTE

sian in order to study it; but there is no evidence that he ever took part
in any Communist activities. Asked by his daughter, about 1922-3, to
what political party he belonged, he replied, ‘To none.’ He did not
agree sufficiently with any. He felt strongly that social reform could
not be achieved by reforming the individual, but only by reforming the
system. His conception of history was materialist: he accepted the
desire for economic gain as the mainspring of human action.

It is not difficult to see how natural such attitudes were to one whose
outlook had been conditioned by the experiences just described. I have
observed elsewhere that, when a child has one parent who is easy-going
and affectionate and another who is severe and apparently unloving, it
identifies itself with the former and is preoccupied throughout life with
its relationship with the latter. When it is the father who is severe, this
seems to lead, in the case of a male child, to a homosexual tendency ;
when it is the mother who is unloving, this leads the male child to a
preoccupation with women and with incest. This was certainly the case
in Briffault’s great contemporary, Havelock Ellis, for instance. It seems
to have been equally true of Briffault.

Richmond , 1958

GORDON RATTRAY TAYLOR

CHAPTER 1

The Origins of Human Society

Our object in this book will be to trace the origin of human society.
What was the character of the first social groupings which arose as
humanity evolved from the level of the animal ? To ask this question
raises the problem of what was the original nature of the family and of
marriage.

Many writers have assumed that the family as we know it — the
patriarchal family — was the first form of human grouping; that societies
were created by the coming together of such families. Professor Maine,
nearly a century ago, put this view forward, and early in the present
century Professor Westermarck restated the idea with a much broader
apparatus of anthropological information. Westermarck went further
and claimed that the family had always been monogamous, polygamy
being a late and decadent manifestation.

But this is by no means the only view of the origin of society which has
been taken. In the year that Maine produced his Primitive Law , the
Swiss jurist Dr Bachofen was writing his Das Mutterrecht , in which he
maintained that the primitive state of man was one of sexual promiscuity
and that the first social groups consisted simply of mothers and their
children without the presence of any father. This view was further
developed by others, notably McLennan, who introduced a novel con-
ception when he claimed that the earliest human groups consisted of
groups of kinsmen having an animal or plant for their badge, a state of
affairs which he called totemism.

Founders of modern social anthropology — notably Morgan, Tylor and
Robertson-Smith — also established the fact that in primitive society the
role of women and their influence was much greater than has been the
case during historical times in civilized societies. This conclusion is
generally called ‘The Theory of Matriarchy’ — a name given to it by
McLennan in opposition to the theory which traced social origins to a
patriarchal age such as is presented in the Bible. The term has been
loosely employed to denote a status of women ranging within wide
limits, from the mere reckoning of descent in the female line instead of
in the male line, to the exercise of supreme authority by women, or
gynaecocracy. It may, I think, be used legitimately in a relative sense,

THE MOTHERS

in opposition to the term ‘patriarchal’, when referring to a society in
which the interests and sentiments connected with women play a more
important part than is the rule in the civilized societies with which we
are most familiar.

Professor Westermarck’s long series of books appeared after those of
McLennan, and at the time of writing the view that society was primi-
tively patriarchal is still widely accepted. In this work we shall attempt
to re-examine the evidence, making use of anthropological material not
available to McLennan.

In our inquiry we shall start from the fact that primitive societies
approximate more closely to animal groups in structure than advanced
ones, and their character is a consequence of this. In animals there is
nothing corresponding to a patriarchal social group — the male has little
or no share in the formation or maintenance of the animal family, and
is often entirely absent from it. If human society developed out of such
groups, it must have had its origin in an association which represented
female instincts only, and human culture must have been moulded in the
first instance not by the fierce passions of hunters battling for food and
women but by the instincts of the mother.

The difficulties of such an inquiry are formidable. The remnants of
uncivilized races at present existing are not truly primitive, inasmuch as
they have survived the tribulations of hundreds of thousands of years.
Even in conditions of isolation and stagnation, time must have deposited
a detritus of culture. Fortunately culture — and I use the term in the
contemporary sense of the inheritance of institutions and patterns of
behaviour, of ideas and beliefs — has its own rules . 1

Firstly, no human sentiment, no idea, no institution, has ever appeared
completely de novo. Secondly, no phase of human ideas, sentiments or
institutions ever fails to leave its imprint upon subsequent phases.
However grotesque some institution, custom or idea, found in some
uncultured tribe, may appear to us, nevertheless its equivalent is
invariably to be found in our own customs, institutions and ideas. The
quaintest cultural peculiarity from the Cannibal Islands or Central
Africa can be matched in modem London.

But despite this undying persistence of primitive forms, it is a strange
fact that primitive behaviour appears to us a fantastic and incoherent

1 Briffault, in fact, does not use the word ‘culture’, which has been popularized
since he wrote, but speaks of ‘traditional heredity’. As the concept of culture is
now much more widely understood, I have substituted the modem term, and
have abbreviated his remarks on it very considerably, but it is only right to
restore the balance a little by stressing that he was one of the very first to
perceive the importance of cultural forces in moulding both behaviour and
personality, and much of what he had to say was, at the time when he wrote it,
of great interest and novelty. — G. R. T.

THE ORIGINS OF HUMAN SOCIETY

product of a mentality utterly different from our own. And despite our
belief that human nature never changes or that it is the same the world
over, many beliefs and customs and institutions which are found to be
common to primitive humanity from the Poles to the Equator, from
Polynesia to the American prairies, nevertheless appear to European
man to be uncouth and strange. To interpret their significance and
origin has been the standing puzzle of anthropology. Many interpreta-
tions have been put forward, but it must be confessed that these ideas
and sentiments still remain to a large extent a riddle.

In attacking this problem, we depend upon two sets of facts. In the
first place man displays a number of instincts which he appears to
inherit and which must represent the result of evolution from animal
instincts. At the same time, he learns from his society much which soon
becomes ‘second nature’; it is clearly important for our purpose to
distinguish very carefully between what is inherited and therefore un-
modifiable, and what is acquired from the culture and thus is peculiar
to a given society. In point of fact, many attributes which are widely
believed even today to be entirely inherited turn out upon examination
to be largely or wholly cultural. In general, the instincts of man appear
to be much less determinate in character than those of animals. With few
exceptions, such as the infant’s instinct to suck, they are quite general in
nature, and their form of expression is largely determined by the
culture . 1

It is often thought that civilized man inherits a higher intellectual
development than the savage, and a greater aptitude for education,
which enables him to acquire more easily than backward races the
cultural inheritance of his society. The facts are, however, that the
children of savage peoples learn just as rapidly as those of Europeans,
or even more so. Thus of the negro children of the coast of Guinea
Captain Binger says: ‘They have an extraordinary memory and capacity
to learn anything that one may teach them. They are quite as highly
gifted as our European children of the same age.’ And a missionary in
East Central Africa writes : ‘It has been the general experience in other
parts of Africa that negro children have no greater difficulty in learning
to read and write than European children, but quite the reverse; and
that experience is confirmed in our own schools.’ When schools were
first established in Hawaii, teachers were often embarrassed by being
unable to keep up with their pupils; while in Samoa the natives deve-
loped such a craze for arithmetical calculation that the Hon. Frederick
Walpole declared that his visit to that island was positively embittered

1 For instance, a tailor bird can only make a particular type of nest. If man
has an instinct to house himself, it is certainly not an instinct to make only one
particular kind of house. — G. R. T.

THE MOTHERS

3 °

by ceaseless multiplication and division. Scientific tests have shown on
a number of occasions that although individuals differ there is no
appreciable difference in the natural mental capacity between savage
races and modern Europeans. It seems, however, to be true that this
aptitude reaches an early ceiling, whereas the European out-distances
the savage in the long run.

It is often supposed that artistic talents are inherited, and attempts have
been made to demonstrate a specific aptitude for music. In particular,
it has been thought that the ability to recognise the pitch of a musical
note when sounded is a rare inborn capacity. However, it has been
shown that almost every child is capable of acquiring it quite easily; it
is only that the majority of children lose their natural musical ability
through lack of training during the most susceptible period.

To many it seems evident that the progress of the human race is
correlated with the development of moral qualities. But Professor Boas
says: ‘As the civilization is higher we assume that the aptitude for
civilization is also higher; and as the aptitude for civilization presumably
depends upon the perfection of the mechanism of body and mind, the
inference is drawn that the white race represents the highest type of
perfection. . . . There is no satisfactory evidence that the effects of
civilization are inherited. , And as Mr Burt says: ‘In inborn mental
constitution the civilized inhabitant of Paris or London today is, if any-
thing, inferior rather than superior to the Athenian at the time of
Pericles or the Englishman at the time of Shakespeare; and indeed, if
anything, inferior rather than superior to his prehistoric ancestors. . . .
The superiority of the modern, civilized man is due, not to the heredi-
tary powers and capacities but to mental contents and achievements
accumulated not by inheritance but by tradition . 5

Social co-operation and other moral factors are often thought to be
inborn, yet many observers have noted how well-conducted young men
can become lazy, disrespectful and dishonest as a result of mental
injuries or even exposure to a different social environment. What are
supposed to be racial characteristics also seem to be derived from
culture rather than inheritance. As Professor Boas says: ‘Much has been
said of the hereditary characters of the Jews, or the Gypsies, or the French
and Irish, but I do not see that the social causes which have moulded the
character of members of these peoples have ever been eliminated satis-
factorily . 5 The French feeling for ‘logic 5 and the English tendency to
blunder through by ‘horse-sense 5 can be deduced from their respective
social histories. The French language itself developed its lucidity and
logical qualities in the French salons and had previously been as un-
couth as English. It is a current view that the acquirements or contents
of the mind are of secondary importance to ‘good stock 5 , and if a Briton

THE ORIGINS OF HUMAN SOCIETY 31

behaves well in difficult circumstances it is often attributed to his
hereditary background or breeding. In much the same way the Gothic
barbarians despised the culture of ancient Europe and refused to have
their children educated because they said ‘education enfeebles the
mind ’. 1

Sir Thomas Browne affirmed the inborn character of all the virtues.
‘Rejoice that thou wert born of honest parents, that honesty, modesty
and veracity lay in the same egg and came into the world with thee/
Such beliefs die hard. The absolute validity of moral ideas is thought to
demand that they should be rooted in the very structure of human
nature. Even social institutions such as private property, monogamic
marriage or male dominance tend to be viewed as embedded in human
nature. Able writers have sought to show that regard for chastity and
the sentiment of modesty are also implanted in the human mind.

The transmission of culture takes place by words, and since animals
have no language there can be no true equivalent of human society in
the animal world. Without language it is impossible to form concepts or
to think in any but the most primitive manner. Thought is but un-
uttered speech ; the infant learns to think and to speak at the same time
— it thinks aloud, and its thoughts take the form of a continuous babble.
Though later the child learns ‘to hold its tongue’, the adult, when no
one else is present, will often talk to himself.

While some germ of thought no doubt exists in the higher animals,
nevertheless, as Professor Lloyd Morgan says, animals are without the
perception of relations; their memory is desultory and ‘the evidence
now before us is not sufficient to justify the prophecies that any animals
have reached that state of evolution in which they are even incipiently
rational’.

Man is man by virtue of language alone . 2 The point is so important
that it is worth a little development. Thus we may draw attention to
the fact that in Latin as in Greek, the same word meant both dumb and
stupid, while our own word dumb derives from the German ‘dumm’ —
stupid. The name of a thing has widely been regarded as intimately
connected with its very nature. Plato’s famous doctrine that a word is
not a mere label or representation but is the true reality, as an indepen-

1 In many cases apparent racial differences are due to malnutrition or endemic
disease. Where aristocratic classes arc of a finer physical and mental type than
the common people, this is probably the result of their privileged position, not
the cause of it.

2 Controversy sometimes arises as to whether thought is possible without
language. The answer is that language is but a particular system of symbolism
and thought is possible by means of any other symbolic system. Mathematicians
can think in terms of mathematical symbols, and the blind deaf-mute Laura
Bridgman was observed to move her hands in sign-language when dreaming.

THE MOTHERS

dent existence, is merely a developed form of a theory found in many
phases of primitive society. The Babylonians believed that the name was
the essence of the person or thing to which it was attached and that that
which had no name did not exist. The Egyptians also held that all things
had their true name — the essence of their being. Thus it is that things can
be created by uttering their names and that creators so create the world.
‘There was a time’, states an ancient Egyptian papyrus, ‘when no one
and nothing existed except the creator. A desire came over him to
create the world, and he carried it into effect by making his mouth utter
his own name as a word of power; and straightway the world and all
therein came into being.’ That creative utterance was known to the early
Sumerian priests as mummu, and in later Babylonian theology as the
God Nebo, called by the Greeks Hermes, who had been identified with
the word as far back as the sixth century. Among the Persians and in the
Vedanta the same conception is found. ‘The world’, says a Vedic
hymn, ‘originates from the word.’ We find the same ideas among rudest
humanity. For example, among the wild tribes of Brazil every creative
thing has as they say ‘its mother’, every lake, every stream, every species
of animal, vegetable has its determinal paradigm or ‘mother’. Father
D ’Acosta, writing of the Caribbean natives, remarks ‘that they approach
somewhat near the proposition of Platoes Idees’. The same comment
was suggested to Father Lafitau when he saw the American Indians.

The person’s name is therefore regarded as his spirit or soul — the
very word soul is from a root denoting wind, and the Greek is

derived from the word Wvxco — I blow.

In many languages the word for breath and spirit are identical.
Hence if one utters one’s own name one is liable to lose one’s soul,
and names are commonly kept secret so that magic cannot be worked
with them. Sneezing, which is an expiration of the breath, is also liable
to cause the soul to depart, and the Malays when they sneeze call back
their soul in a loud voice.

It is by the same conception of the importance of the word that
spells, curses and blessings and incantations operate. The Bechuana,
when they go to war, have a priesdy personage who marches with the
troops and continually calls out ‘the army is not seen’, which effectually
conceals the troops from the enemy. No sorcerer can harm a person if
he is ignorant of his name, and an Irish poet quite failed to curse a
certain King of Ulster whose name would not scan in any known metre.
Accordingly, to work magic it is necessary to know the correct words
and in Persia the magic art was called ‘The science of names*. All that
Eskimo require to command the spirits is a knowledge of their
names. The power which such knowledge bestows is illimitable.
Australian medicine-men can stop the sun, cause thunder, raise moun-

THE ORIGINS OF HUMAN SOCIETY 33

tains, create lakes and rivers; in Brazil they can control the course of
nature and human life.

By the same argument, poets are creators and even sorcerers. ‘None
merits the name of creator save God only and the poet’, said Tasso.
The word magic is derived from the word for ‘making* and vates
means both prophet and poet. The word runa meant first of all the poet’s
magic word and then poetry in general. The classical invocation ‘Sing,
heavenly muse!* was originally the formula by which the poet invoked
the god who possessed him. The Hindus and Arabs have a simple
means of depriving a sorceress of the power to do harm : they extract her
front teeth, so that she is unable to sing or articulate distinctly; and it
was for analogous reasons that Odysseus had the power to tame animals,
heal sickness and compel the affection of women. Vergil assures us that
poetry can draw down the moon from heaven; while the ancient Irish
were convinced that a satire inevitably caused blisters to appear on the
face of the victim.

It follows that to doubt that language is innate in every human being
when born is tantamount to denying the existence of the soul. Eunomius
charged St Basil with atheism for maintaining that babies learn to talk.
Many rulers have sought to put the matter to the test: James IV of
Scotland put two bairns in charge of a dumb woman on the Island of
Inchkeith and we are assured that when they came to the age of speech
‘they spak guid Hebrew* — for Hebrew was, of course, the primitive
language of the world. The view that language was given to man and
not invented by him was maintained even as late as the nineteenth
century by the famous Catholic philosopher Vicomte de Bonald.

But in fact the numerous discoveries of children who have been
brought up wild by animals shows that in no case could they speak,
nor could they even think. Madamoiselle Leblanc, the wild girl found
near Chalons in 1731, said that in her pre-linguistic days she had no
thought at all and had now only very dim memories of the period.
Kaspar Hauser, the youth found in 1828 near Nuremberg, had the
mind of a child and mistook inanimate objects for living things. It
would appear that the same is true even of animals — a cow which is
born deaf will not low, and sparrows brought up with canaries have
acquired the trills and call-notes of the latter. The paradox is that
human culture must be acquired in order to be inherited — a paradox
which Goethe expressed by saying, ‘That which from thy fathers hast
inherited, acquire it! — that it may be thine.* Nevertheless, the belief
in inborn character dies hard.

We may see in the most ordinary child evidence of the primitive
character of the human animal before it has become influenced by
culture. The child is amoral ; it is callous, if not cruel ; it is attracted by

T.M. — 2

THE MOTHERS

dirt and those things which we regard as obscene; it is selfish, passion-
ate, gives way to anger and resents deprivation bitterly. In an adult
such behaviour would be regarded as insane, and it is hardly too much
to say that we derive our sanity from the culture. It is this socially and
traditionally acquired mind which thrusts our physiologically inherited
attributes into the obscurity of unconsciousness, and which may be
equated with Freud’s super ego and id.

In short, all those mental characters which are specifically human
are derived from the culture. They include not only the conceptual
forms of consciousness — thought and ideation, but also those feelings,
sentiments and effective values by which the behaviour of the social
individual is for the most part determined. The human mind, in all
that distinguishes it from the mind of animals, is thus a social product,
and it is for this reason that I have already said that there is nothing
properly corresponding to a human society in the animal world. Thus
it is in the social history of the human race itself that the origin and
development of the human mind — in so far as it is human at all — are
to be sought. The true field of investigation into the psychology of
mind is that of social anthropology and social history.

CHAPTER 2

Motherhood

The nearest thing to an emotional relationship among animals, and
the presumable point of origin of all social relationships and structures,
is that which subsists between some primate animals — mother animals
and their offspring. If, as I hope to show, it is from this germ that
human society sprang, we must start by asking how the maternal
instinct originated.

As we ascend the evolutionary scale, the role of mother becomes
increasingly important, but let us take the argument back a stage
further and ask how it is that mothers come to exist at all. In the simplest
forms of life reproduction can take place asexually. In many higher
forms reproduction is effected by the female alone parthenogenetically. 1

Sexual differentiation is not rooted in the constitution of life, but is
a biological accident, a special device designed to meet certain condi-
tions. This differentiation becomes marked in the higher organisms.
Whether males or females are produced of such unions depends
essentially upon nutrition — good food giving rise to an excess of females.

While the function of the primitive female is fulfilled when she has
produced an ovum, as we advance up the evolutionary scale another
duty becomes thrust upon her — that of care for the offspring. In the
simplest stages this care is perhaps involuntary. The young of some
animals attach themselves parasitically to their mother and suck the
fluids of her body. In the case of some gall-flies the young actually eat
their mother, and creep out of her empty skin only after the matricidal
feast is over.

As we ascend the evolutionary scale we find that the period allowed

1 This is the ordinary mode of reproduction among wheel-animalcules and is
common among many species of crustaceans. It has been observed among thrips
and nematode worms, and is the rule among saw-flies and caddis-flies. ‘Among
gall-flies the male is useless; the continuation of the species being effected by
virgin females, although males exist.’ It commonly takes place among several
species of moths and butterflies. Reproduction of males without the female is,
however, unheard of. In some species which habitually reproduce partheno-
genetically males may be entirely absent, as for example in Cypris ovum and
other ostracod crustaceans and in fresh-water rotifers. In the Solenobia butterfly,
males appear only occasionally, sometimes at intervals of years. In some species
of rotifers the males are not sexually functional.

THE MOTHERS

for gestation steadily increases in length. It is true that an elephant
carries its young for two years but, proportionately to weight, human
beings carry their young longer than any other animal. This increased
period of gestation allows for an increasingly complex development in
the organism, and is generally associated with the reduction in the
number of young produced. Furthermore, the young require increas-
ing amount of care after birth in the highest stages of development.
Rats, for instance, shift for themselves when thirty-nine days old and
reach maturity in six months; a hartebeeste antelope, one week old,
can outrun the fastest man. A young elephant is capable of following
its mother when two days old. Carnivorous animals, on the other hand,
are bom helpless; they are unable to stand for several days, and are
entirely dependent upon their mother for many months. Young lions
are unable to stalk for themselves until they are about a year and a half
old. In the case of monkeys the period of care is even longer; a baby
gibbon remains clinging to the body of its mother for seven months;
the orang-outang at the end of one month is as advanced as the human
baby when a year old; a lamb a day old has proceeded farther in its
development than either. This period of immaturity is not required
for bodily growth but for completion of the brain and neural system.

Similar (differences in rapidity of growth and maturation can be seen
between the lower and more civilized races of men; savage children
develop more quickly than Europeans but, their development being
completed earlier, are less capable of further progress. For example,
children among the Baholoholo of the Congo know how to paddle a
canoe and catch a fox at an age when civilized children are still in the
arms of their nurses. Among the Kirghis a child of three can already
ride a horse and at six takes charge of a herd of camels. Among the
Habbe of Nigeria children of six or eight leave their home, build a hut
and provide for themselves. Among the Aleuts, children of ten have
already become hunters and not infrequently keep a wife. Chiriguano
children of seven or ten go on war and hunting expeditions with their
fathers. Unfortunately mental development ceases about the time that
sexual maturity is reached. Captain Binger remarks: ‘Not only does the
intellect of the child cease to develop, but it might be said that it
retrogresses; the memory becomes impaired. He becomes stupid, mis-
trustful, vain and deceitful as he was formerly intelligent.’

Thus it would appear that the congenital superiority of what are
regarded as the higher races of man consists essentially in a slower rate
of development, and this slower development is rendered possible, and
perhaps caused, by the prolonged relief from the necessity of providing
for subsistence and protection — in a word, by mother-care.

But the mother animal’s instinct to look after her children is, in ani-

MOTHERHOOD

mals, of a purely physiological kind. It is not something which is always
with her, but comes into being only at the time of pregnancy or just
before birth. Unimpregnated female mice take no notice whatever of
young mice, and it is not until the later phases of pregnancy that they
will lick them and endeavour to carry them away. Manifestations of this
instinct cease about six weeks after birth ; moreover, it is often satisfied
by substitutes — a pregnant bitch, noted Loisel, searched restlessly until
she was presented with a litter of young rabbits, when she was entirely
satisfied and licked and fondled them. A female crab bristles with
anger if the maturing larvae on its appendages are touched — behaviour
which would provide a text for maternal devotion — but the crab be-
haves in just the same manner when the brood-stalks have been appro-
priated by noxious parasites.

Maternal reactions generally take place in response to gross physical
stimuli. Suckling is sought by the female to relieve the tension in the
mammary glands, and brooding in birds is caused by exhaustion and
irritative congestion of the abdominal wall. In birds that are not good
sitters, the defect may be remedied by rubbing their abdominal skin
with nettles. Consequently when we speak of maternal 'care’ and
‘devotion’ in the lower animals, we are really projecting human senti-
ments into behaviour which is in fact purely physiological.

That the maternal instinct is a product of evolution and not of
primary impulse is indicated by the fact that it requires an appreciable
time to develop fully. The same females that will offer their life for their
young will commonly eat them when they are new-born. Carnivores
are prone to eat their young whenever they are frightened, as are sows,
and even dogs and cats, while the reindeer is said invariably to kill its
second fawn. Even in humans mothers not uncommonly take a day or
two until they become attached to their children, and at the moment
immediately following birth infanticide is common among both savage
and civilized mothers, whereas a little later it would be difficult or im-
possible.

Maternal instinct is also frequently limited in duration. Few animals
pay regard to earlier offspring after the birth of fresh young ones, and
many animals throw out the young after lactation has finished. Birds,
for instance, change from ‘unceasing solicitude to open hostility’ and
drive forth their young from the neighbourhood. Swallows and house-
martins when migrating frequently abandon their nestlings. While
carnivores frequently defend their brood, the practice varies — for
example, the offspring of a seal can be killed before its mother’s eyes
without causing her the slightest concern, whereas the walrus mother
will fight to the death in defence of her young. Nevertheless, on maturity
the maternal instinct ceases.

THE MOTHERS

Among monkeys and apes, however, the maternal instinct is so intense
and uniform as to contrast with all other mammals. The tense and watch-
ful anxiety of the monkey mother has often been noted. Baboons will
wash their young in a stream, and the Cebus has been seen to drive
away the flies which plagued its infant. Some monkey mothers will die
from grief when deprived of their young, and there are several cases
on record of a monkey placing its body between its child and the
hunters’ gun and so sacrificing its life for its offspring. In contrast,
carnivores, although they may fiercely defend their young, will occasion-
ally abandon them. Starving tigresses have been known to kill and eat
their young — behaviour which has never been reported of monkeys.

The new relations which are established by protracted mother-care
transform the very springs of action and behaviour, for they cause the
individual mind to be linked up with others to form a new organism —
the group bound by social ties. Let us consider next what we can infer
from animal societies about the origins of social feeling and gregarious-
ness and the character of the primordial family.

CHAPTER 3

The Origin of Social Feeling

I t has been generally assumed that feelings of tenderness and affection
are part and parcel of the attraction between the sexes, an attraction
commonly spoken of as love. Love is generally identified with the
sexual impulse, and is assumed to be one of the mainsprings of the
universe. Life, says Schiller, is ruled by Hunger and Love.

But love, far from being a basic force in the universe, has arisen late
in human evolution. Primitively, sexual feelings are associated not with
tenderness but with delight in the infliction of pain. Animals are preying
beings and the sight of a mangled or a weak and helpless creature
means to them food. All carnivores are cannibalistic; lions frequently
eat their mates or their cubs. A jaguar in the New York Zoo showed
every sign of fondness for a female kept in an adjacent cage. He purred,
licked the female’s paws, and behaved like the most love-sick admirer,
but when the partition between them was removed, his first act was to
kill her. Sexual hunger, as it is aptly called, is a form of voracity. The
object of the male cell in seeking conjunction with the female cell is
primarily to improve its own nutrition. The female requires the male
also to assure her nutrition, and in some forms of life the female devours
and assimilates the male.

With both sexes, sexual attraction is pre-eminently sadistic, and is
gratified by pain. Rutting animals maul each other, and emerge from
the sexual combat bleeding and mangled. Crustaceans usually lose a
limb or two in the encounter. All mammals, without exception, use
their teeth on these occasions. So fundamental is the association between
sex and cruelty that it is probably never wholly absent even in human-
ity. According to M. d’Enjoy, the kiss has developed out of the love-
bite. Among the ancient Egyptians the word translated by ‘to kiss’
meant ‘to eat’.

Sentiments of tenderness between the sexes are originally connected
not with the sexual impulse but with the mating instinct — something
very different. Mating, meaning association between the sexes, is an
adaptation to the reproductive functions of the female. With the exten-
sion of maternal care, the female needs protection and help in procuring
food, and so requires the co-operation of the male. For birds, where the

THE MOTHERS

female sits on eggs, such help is almost indispensable ; hence it is among
nidicolous birds that the mating instinct has achieved its most striking
development, but even here it is confined to those species which hatch
their eggs by prolonged brooding.

Many mammals, far from associating after the sexual act, appear to
be actually repelled from one another. In general, mammals do not
mate, but the sexes roam in separate herds. This is true of all herbivores.
Elephant cows, after being impregnated, form bands, from which males
are driven off, as do seals and walruses, bears and leopards. Among
carnivores, the female generally conceals herself and her brood from
the male lest he eat the cubs. Even where an association exists between
the parents, it is the female which attends to the feeding of the young.
The male lion is frequently represented as bringing his kill to the female
while she remains with her cubs, but in fact he drags his kill to his lair,
whether there are cubs or not. Not only is it the female alone who pro-
tects the offspring, but in some cases she protects the male also, as
among deer and antelope, and also elephants.

Among primitive human races the mating instinct is almost equally
unimportant. As will later be shown, cohabitation is very transient in
the lower phases of human culture. Conjugal affection is almost un-
known. ‘If you wish to excite laughter,’ says Father Petitot, ‘speak to
the Dene of conjugal affection.’ Similar reports come from many parts
of the world — from Africa, New Zealand, Papua and elsewhere. The
point has been the subject of controversy, but the data which will be
brought forward later in this book will show that in primitive peoples,
sexual love, as we conceive it, is at best rudimentary.

On the other hand, there is no doubt that among savages the love
of mothers for their offspring is unambiguous. In the very peoples who
are described as being devoid of love between man and woman, the
strength of maternal affection is constantly noted. For example, among
the Eskimo the coldness of the sexual relation is conspicuous, but
maternal love is said to be ‘lively and tender’, more so, as Father
Petitot remarks, than among ourselves. Among the Dene ‘maternal love
is developed to the point of obliterating every suggestion of prudence,
and even every reasoned act of intelligence.’ The Patagonians display
a love for their children ‘which is quite extravagant; they show such
extreme compliance with regard to them that whole tribes have been
known to leave a district or to remain there longer than was advisable
simply to gratify the whim of a child. Primitive women will court
danger or pain for their children. During a tribal war in Samoa ‘a
woman allowed herself to be hacked from head to foot, bending over
her son to save his life’. Women will not eat until their children have
been satisfied; they offer themselves to redeem children who have been

THE ORIGIN OF SOCIAL FEELING 41

taken as slaves. This intensity of maternal affection is equally noted
in the lowest races, such as the Fuegians, the Seri Indians, the
Veddahs, the Ainu and others.

It might seem that the widespread practice of infanticide was irre-
concilable with the existence of maternal love, but it must be recalled
that in primitive society murder is often not a crime, and it is believed
that the dead are still with us. Thus in the Cameroons, during the
German invasion, the natives, who are devoted parents, killed most of
the new-born, ‘in pity for their sufferings and in the firm belief that
their spirits would return to earth as soon as all was peaceful once more’.
In primitive society corporal punishment of children is unheard of.
Father Le Jeune complained that ‘all the savage tribes of these parts, and
those of Brazil, as we are assured, cannot chastise a child or bear to see
one chastised. What trouble this will cause us in carrying out our inten-
tion of instructing their young!’ In fact, many missionaries have stirred
up trouble by their chastisements.

In other respects, the primitive mother is far from gentle. Her feeling
for her mate is primarily that of loyalty. She is content to leave the choice
of mate to her parents, and her estimate of the husband’s value is a
hundred per cent utilitarian. His appeal therefore grows with the close-
ness and stability of his relationship with her. Among the lower races,
where sexual selection exists, it is predominantly exercised by the
women, as we shall see in detail later. In those races where the men are
indifferent or brutal to the women, strong and genuine attachment to
them is nevertheless displayed, and their widows are inconsolable
when their brutal husbands die. Fijian women, who are among the
most brutally treated, insist upon being killed on the graves of their
husbands. It is probable that the custom of ‘suttee’ was originally
voluntary. The numerous wives of an African chief, whom he uses as
pillows and footstools, vie for the honour of being so employed.

The female’s affection provides a powerful inducement for the male
to modify his sexual instincts to suit her mating needs, for her affection
is the equivalent of the maternal tenderness which he knew in infancy.
Biologically, however, his sole function is the impregnation of the ova,
and his instincts are limited to that. These instincts are indeed in
antagonism with the mating instincts. The sadistic hunger of the mascu-
line impulse can never become entirely blended with mating attraction.
T1 n’y a rien de si loin de la volupte que rattendrissement,’ observes
Lamartine. Indeed, the two forms of sexual attraction are essentially
incompatible. Love — tender feeling — is a common cause of ‘psychical
impotence’. The two instincts, sexual and the mating, may exist in the
male quite independently, and this, as we shall see, is common in
primitive humanity. The sexual impulse may have no trace of affection,

T,M. — 2*

THE MOTHERS

while genuine attachment may be unattended by sexual jealousy.
Throughout primitive societies the distinction between the two func-
tions is much more consciously recognized than in our own, where
sentiments and institutions have tended to obliterate the distinction.

The moral qualities, such as courage and character, which are supreme
in the woman’s sexual choice have no place in the man’s in so far that
the choice is purely sexual; he discriminates chiefly with regard to sexual
values; that is, the physical qualities of youth and beauty, which are
ultimately expressions of the suitability of the female for rearing the best
offspring. Hence the feminine taunt that a man may be attracted by a
woman whom he neither esteems nor respects.

Since the female instinct is more directly founded on biological
need, and is developmentally older than the male, it tends to retain its
primitive character, even in advanced stages of culture. This is why it
is less subject to cultural transformations in women than in men. Even
a cursory study shows what a wide range of fashions in love have existed,
and it has frequently been remarked that romantic love is profoundly
influenced by literature and traditions, and that no one would be sub-
ject to it in the same form had he never read a novel nor seen a play.

Yet people constantly treat animal and primitive sex impulses as
identical with the love of cultured humanity. Sexual love is spoken of
as if it were a simple basic emotion or impulse, when it is really the most
complex of sentiments. It is compounded of self-esteem — ‘amor a
nullo amato amar perdona’, of aesthetic elements — few men would be
attracted by a woman who was a habitual frump or slut — and, by
participation in common tastes, mostly imaginary. As Spencer says,
love ‘fuses into one immense aggregate most of the elementary excitations
of which we are capable’.

Just as the transferred affection of the female for the male is a deriva-
tive of maternal love, so all feelings of a tender, altruistic character are
extensions and transformations of the maternal instinct. Altruism is
almost unknown in animals, and is in contrast to basic biological de-
mands. These sentiments, being of female origin, develop more strongly
in the female. The sympathetic, protective, affectionate attitude, trans-
ferred by the female from the offspring to its father, is gradually ex-
tended towards all males, towards females who are not possible rivals,
towards animals and all living things, and even towards plants, flowers
and inanimate objects. These are handled tenderly, whereas the male is
disposed to be rough and to destroy. For the male the extension of
sentiments in this way is much more difficult, more unnatural. The
male child is born cruel. It is his natural propensity to inflict suffering
and to destroy. Only social education can develop a tender disposition
in him to any degree.

THE ORIGIN OF SOCIAL FEELING 43

Long before such a transformation occurs maternal instincts produce
an even stronger bond of attachment in their direct object, the offspring.
The strong feeling of the child for his mother, founded upon infantile
experience, is highly developed in primitive societies, but is weakened
in advanced societies where sentiments of manliness and independence
are stressed . 1 Savages remain children.

The primitive’s continued dependence upon the mother’s affection
is a feature of primitive psychology. The Indians of California ‘scarcely
acknowledge their father but they preserve a longer attachment for their
mother who has brought them up with extreme tenderness’. Among the
Iroquois ‘the crime which is regarded as most horrible and which is
without example is that a son should be rebellious towards his mother’.
In Melanesia, when engaging a boat’s crew, one comes upon men of
forty who say that they must first obtain their mother’s consent. The
Dayaks of northern Borneo are devoted to their mothers and honour
them all their lives, but ‘their father they may like or they may not ; they
recognize no duty towards him; but their mother is something holy to
them, whatever she is like, and no one is ever allowed to breathe a word
against her’. Similar examples are found in almost every part of the
world. The strongest of all natural ties, says Wilson of the West African
negroes, ‘are those between the mother and her children. Whatever
other estimate we may form of the African, we may not doubt his love
for his mother. Her name, whether dead or alive, is always on his lips
and in his heart.’

Daughters, too, are more closely bound to their mother than to their
husband. In Togoland, ‘the bond between mother and daughter is so
strong that both remain bound to one another until one dies. Never can
love towards the husband displace in the heart of a daughter the love
towards her mother’.

The attachment of the young towards the mother consists, not so
much in the sentiment of tenderness, as in the sense of dependence and
fear when protection is withdrawn. The child is therefore ready to
accept a substitute protector. All young animals will attach themselves
to the first creature, animal or human, that will look after them. Mr
Selous mentions that, having shot a female rhinoceros which had just
dropped a calf, the latter at once trotted behind its mother’s slayer, and
quietly followed him to his camp. (Such instances show how animals
were first domesticated.) The reliance upon the mother extends to all
companions who are recognized as not being hostile or dangerous, and
results in the general disposition to friendliness.

Brothers and sisters are naturally the first substitutes adopted to

1 In some tribes boys are taught to use violence towards their mothers in
order that they should grow up manly.

THE MOTHERS

satisfy this desire for dependence and to appease the fear of solitude.
Such feelings are even more prone to assume the character of affection
when directed towards companions of the same age than when directed
to the mother. An instinct of clannishness becomes a marked feature of
such a group.

It is often held that gregariousness is a fundamental instinct. In fact,
it is evolved from family feeling. The majority of animals are not
gregarious; in those that do form herds, the herds are constituted of
family groups. Far from there being any natural tendency to congregate,
most living organisms tend to disperse, for there are biological advantages
in wandering to fresh fields and pastures new, and concentration tends
to exhaust the food supply. Primitives, far from displaying signs of any
general social instinct, exhibit profound distrust and hostility towards
those who are not members of the group. ‘In primitive culture’, observes
Dr Brinton, ‘there is a dual system of morals; the one of kindness, love,
help and peace, applicable to the members of our own clan ... the other
of robbery, hatred, enmity and murder, to be practised against all the
rest of the world.’ It is for this reason that small groups are so unwilling
to merge, and that tribes in the same territory are deeply suspicious, and
their languages diverge so far that they cannot converse with one an-
other.

Wherever groups exist, it is as the result of specific needs and instincts,
not the result of any generalized mutual attraction. In the higher forms
of animal life the so-called social instinct is the direct outcome of the
relation between mother and offspring, and of the reflection of maternal
instincts between members of the same brood or brotherhood, as
Darwin perceived. The material out of which all human society has been
constructed is the bond of those sentiments. The origin of all social
bonds, the only one which exists among the higher animals and in the
most primitive human groups, is that created by mother-love.

CHAPTER 4

The Herd and the Family
amongst Animals

Let us now consider animal societies and herds.

Supposed Animal Societies

Every association of individuals in the animal kingdom is, without
exception, a reproductive organization, even though such groups often
serve other purposes, such as mutual protection or the securing of
food. 1 True, in insect communities the correlation which takes place
between the male and the female is elaborated so that a large number of
differentiated individuals co-operate to achieve the reproductive end,
but the principle is the same. Insects are, however, an offshoot from the
main stem of evolution, and lie outside the direct line of human ancestry.

Many visible assemblages of animals are not true associations. Fishes
collect in shoals simply because they seek the most favourable conditions ;
just as sea-birds congregate on rocks. Flocks formed for migration are of
analogous origin, and revert to separate groupings as soon as the migra-
tion is accomplished. The same is true of deer and antelope. Sometimes
the presence of a large number of species, such as African reed-bucks, in
one place, gives rise to the impression that they are gregarious.

The association of lions in groups of as many as forty is, as Selous
remarks, only of a very temporary nature. Hyenas, too, are solitary
animals, even though they are often seen in packs, since they scent the
blood of a dead animal from afar and collect for the feast. Wolf packs are
the nearest approach to an organized association, but they are not per-
manent associations, and the South American wolf never forms packs.
The ‘pack law’ of the Indian wolf, so eloquently described by the author
of The Jungle Book , exists only in imagination; it is scarcely ever known

1 The tendency to interpret biological phenomena in terms of human tradi-
tions and assumptions is illustrated by the conceptions put forward at different
times about bees. The ancients regarded the egg-laying female as a patriarchal
male, and called it ‘the king’. The hive was divided into patricians and plebians.
When the true sexes of the insects became known, the female was called ‘the
queen*, and the hive was regarded as a more democratic organization, and com-
pared to an industrial community under the term, ‘hive of industry.*

THE MOTHERS

to gather in larger numbers than three. Another animal frequently
spoken of as organized into communities is the beaver, and beaver
colonies have been described as working under a leader. Actually
beavers live in families and have no leader. The largest communities
ever seen by Agassiz consisted of five families. Tt is evident’, he says,
‘that beavers are not really gregarious in their habits, and that the dams
and canals are the work of a comparatively small number of animals.’
A dam has been known to be constructed by a solitary beaver. Equally
romantic descriptions have been given of the ‘societies’ of ‘prairie dogs’.
According to Captain Marryat, their burrows are laid out in regular
streets, forming prairie-dog cities. They are in the habit of paying
afternoon calls on one another. It is scarcely necessary to say that this is
pure fable.

Polygamy and Monogamy amongst Animals

Are any animals ‘monogamous’ ? Woods Hutchinson has made the
extraordinary claim that among animals ‘monogamous marriage lasting
for life’ has ‘become adopted by every dominant race on account of its
resulting in the largest number of most efficient offspring’. Several
writers have been anxious to discover such a relationship, and to suggest
that ideas of sexual morality which developed at a late stage in human
history have some sort of validity among the lower animals. The facts
show, however, that whether such relations are polygamous or mono-
gamous has no bearing on reproductivity. Actually the terms, ‘mono-
gamy’ and ‘polygamy’, applied to animals, do not have the same mean-
ing as when applied to human marriage, for they refer only to the
associations which take place during one breeding season. The so-
called ‘true marriage’ of birds merely means pairing during a single
season. Thus the mallard, while essentially monogamous during the
pairing period, as soon as the duties of incubation have begun, savagely
pursues every other female that ventures in his neighbourhood. Simi-
larly, the bull moose, which is spoken of as monogamous because it is
rarely found with more than one cow at a time, does not remain with
her for more than a week, and, since the rutting season lasts about two
months, he mates with a considerable number. The same is probably
true of the majority of animals which, though not forming polygamous
herds, do not remain with the female after sexual union. The lion is no
more monogamous than the domestic cat. Among mammals ‘a union in
pairs lasting beyond one season’, say Brehm’s editors, ‘has been ob-
served only in the case of the dwarf antelope . . .’, in which species
pairing depends upon the circumstance that two young, a male and a
female, are generally born together, and these subsequently pair, the
species being perpetuated incestuously.

THE HERD AND THE FAMILY AMONGST ANIMALS 47

While monogamous associations are found among the lower mam-
mals, such as the rodents, polygamy is universal amongst the highest
mammals, the monkeys and the anthropoids. A report, often quoted,
that some species of monkeys are monogamous seems wholly unsub-
stantiated. Dr Westermarck quotes the assertion of Dr Hartmann that
‘the gorilla is monogamous’ as a foundation for his theory of human
marriage. It is quite incorrect. Gorillas live in groups polygamously, as
numerous observers have reported. Hartmann bases his assertion on an
article in a popular magazine, and this, when inspected, proves not in
fact to make such a statement.

Jealousy among Animals

If the patriarchal group derives from animal society, we should expect
animals to display sexual jealousy. Among carnivores, competition
among males for the possession of females is much rarer than in the
herbivores, and it is the latter which are equipped with weapons of
sexual combat. Lions never fight; bats queue quietly for their turn,
while other males pair with the female. Though deer fight, they do so
only during rut, and this combativeness does not denote anything cor-
responding to jealousy, which, as we use the term, implies the choice of
a particular female, the appreciation of differences. The displays of male
decoration serve to attract attention, as do call-notes, but not to influence
individual choice. Some animals, such as salmon, fight in order to
protect their eggs, as does the stickleback, whose combats are so furious
that they have often been described in terms of human sentiments, but
the stickleback is by no means monogamous, as was once believed, but
endeavours to induce a number of females to deposit eggs in the nest he
has built.

The struggles of male animals are not for the possession of particular
females, but for access to females in general, and commonly take place
when no females are actually present. Male animals fight for the oppor-
tunity of reproduction, as they fight for food. Brehm placed two pairs
of bears, which appeared to be affectionate couples, in the same pit.
The males immediately fought for the possession of both the females at
the same time, and the winner immediately paired with the she-bear of
the loser. The carnivora are jealous of all rivals, and imprisoned lions
are even jealous of the keeper who approaches the cage. Stags often herd
their does and force hinds to stay in the herd, but they have often been
seen to leave the whole herd and take possession of a new one ; some-
times to exchange herds. Often several stags keep possession of a herd
together, and fights are by no means as frequent as some people believe.
Though seals fight, the bull does not appear to know his own females
individually. Once one has strayed outside his marked-out territory, he

THE MOTHERS

pays no further attention to her. ‘It is the cows and not the bulls’, says
Major Hamilton, ‘which have the real control of the harem system.’
Some young bulls were even observed to mate with females under the
eyes of the old bull without his paying any attention. The male becomes
spent after a short period of rut and loses interest.

The Female in the Animal Group

Though it is the male which keeps the females in a polygamous herd
together, his control is purely in sexual matters. An animal group is not
usually led by a patriarch — an old male — but by a female, who guides it
and keeps watch against possible dangers. Whereas stags will often walk
blindly into danger, the females seem more alert; the same is true of
buffaloes, of African antelopes, of gazelles, chamois, zebras and so on.
‘Sometimes’, says Dr Hornaday, ‘a herd of elk is completely tyrannized
by an old doe, who makes the young bucks fly from her in terror, when
one prod of their sharp antlers would quickly send her to the rear.’
When a male and female tiger are found in company ‘the tigress is
generally in advance of the male’; and the same is true of the lioness.

The migrations of mammals are led by the females, and appear to be
determined mainly by the females’ search for favourable grounds to
bring up their offspring. In short, even in the typically polygamous herd,
the dominance of the male is qualified, and the protective social functions
are discharged chiefly by the female. It is in the field of sexual relations,
in the narrowest sense, that the herd is dominated by the male, and then
only during the period of rut. Thus, the real animal group depends on
the bond between mother and offspring, and not upon the association
between the sexes. Indeed, the male sexual instincts tend to break up
the herd. In the typical herbivorous species — for instance, the red deer
— the female withdraws from the herd after impregnation, and leads a
solitary life with her calf until the autumn. If the hind then joins a herd,
her male offspring is driven off by the stag, and is thus weaned, although
a female calf is allowed to remain.

In short, there is opposition between the maternal and the sexual
instinct. The tendency of the female is to segregate herself with her
offspring, and the more prolonged the maturity of the offspring the
more complete will that segregation be. Since the herbivorous mother
neither feeds nor trains her young, maternal instincts must be regarded
as more highly developed in the carnivores than in the herbivores. Thus
we have the paradox, as it seems to us, that those animals termed
gregarious are devoid of any binding social sentiment, and it is among
the supposedly non-gregarious creatures that the germs of the social
instinct have developed. It is in the maternal not the sexual association
that the growth of the so-called ‘social instinct’ takes place.

THE HERD AND THE FAMILY AMONGST ANIMALS 49

The mother is the centre and bond of the animal family. When the
male attaches himself to it, his association with it is loose and precarious,
and in no animal species docs it appear to survive the exercise of the
sexual function. Nor is any tendency towards a closer association of the
male with the family discernible as we ascend the scale of evolution.

It is usual to apply the term ‘family’ to animal groups, but there is, in
fact, no analogy between the animal family and the patriarchal human
family. The former is entirely the product of the female’s instincts, and
she, not the male, is the head. The male may be more powerful, though
this is not invariably so. But, in any case, this has nothing to do with
his relation to the female. Nowhere do we know of the male using com-
pulsion towards the female. Nor is the position of the male in the animal
group affected by mental superiority, for in animals, whatever may be
the case with humans, it is clear that the female is superior. Masculine
intellectual superiority, if we assume it, is the product of advanced
social evolution, and has no application in primitive conditions. The
female animal is the more cautious, ingenious and sagacious, the male
is reckless and often stupid. With most animals males are much more
often caught or shot than females. It has been said that monkeys, how-
ever, are led by old males. If this is true, it may be due to the females
being burdened with their young, which they carry in their arms. In a
troop of chimpanzees observed in a reserve in Teneriffe the group was
led by a male, but the rearguard was brought up by an old female, to
whom the group ran in time of danger, and who seemed to carry the
group with her when she changed her place.

The female, not the male, determines the condition of the animal
family, and when she can derive no benefit from association with the
male, no association takes place. It is the female which does the house-
hunting; which, in the case of beavers, builds the dams or makes the
burrows. It is the female bird which builds the nest.

While the female instinct produces the family, the masculine in-
stinct produces the herd, for the masculine instinct compels the male
to impregnate as many females as possible. The stability of the herd is,
therefore, limited by the operation of the sexual instinct. It is not an
assemblage differing merely in size from the family, but is opposite in
kind, and, when true herd association takes place, means the dissolution
of the maternal group or family. Thus the herd results from atrophy of
the maternal instinct. As Mr Pycraft observes of animals, this ‘poly-
gamy* arises not because the males capture the females but because the
females seek the males in their desire to satisfy their natural craving.
This extinguishes any feelings of jealousy.

From these facts, it can be understood that the masculine herd is
patrilocal; that is, the female follows the male’s place of residence.

THE MOTHERS

whereas the animal family is matrilocal; that is, the males goes to the
female’s lair. Wherever circumstances bring together a large number of
animals, there is a tendency for the family to lapse into the herd.
Rodents, such as rats or rabbits, which gather in large numbers,
become promiscuous. It is, on the other hand, well-nigh impossible
for the herd to become transformed into a group of families except by
breaking up completely.

CHAPTER 5

The Primitive Human Group

What was the form of the most primitive human groups? This is a
question of the first importance, for upon the answer all subsequent
human development, social and mental, has depended. Presumably
the earliest human societies developed out of some form of animal as-
semblage. They were, therefore, like all animal groups, primarily
reproductive in function, and not, like existing human societies, co-
operative organisations.

Certainly the human group did not develop out of the animal herd
for, as we have seen, the herd is incompatible with prolonged maternal
function. It must therefore have been derived from the animal family
based on the maternal instincts of the female.

At the same time it is impossible to suppose that nascent human
society consisted of small isolated groups corresponding in size to what
we understand by families. Such limited groups could offer no scope
for the development of those distinctively human social relations and
feelings which are essential for the mental evolution which took place.

Some races on the lowest level of culture are, it is true, found in
small scattered groups which include sexual societies and may be called
families — for instance, the savages of Tierra del Fuego, the Eskimo
and some of the most wretched tribes in the forests of the upper Amazon
and Parana. The forest Veddahs of Ceylon are also found in small
groups, consisting for the most part of sexual partners, or single
families.

Such populations are not simply ‘primitive’ in the sense of having
remained wretchedly backward ; they are also unsuccessful races which
have been driven towards the least habitable parts of the earth. The
Fuegians, who belong to the same stock as the other races of South
America, have been ‘pushed off the edge of the world’, and ‘forced to
break up into small clan or family groups’. The forest tribes of Brazil
have been driven to their present habitat by Caribbean invaders. In
these unfavourable conditions ‘it is safe to assume that they never
could have emerged from their savage state’. Likewise there are clear
traces among the Eskimo of a former organization into clans, now
fallen, owing to physical circumstances, into almost complete decay.

THE MOTHERS

The Veddahs of Ceylon are the descendants of the aboriginal royal race
who were driven into the jungle by invaders.

In short, such societies represent, not the primitive condition of
human society, but the effects of its dissolution under the pressure of
unfavourable conditions; we cannot infer from them the condition of
incipient humanity. They arc defeated unsuccessful races; the animal
race which developed into humanity must have been a favoured race
living in exceptionally propitious conditions.

All that is involved in human evolution postulates a much larger
group than the family — and, in fact, we find all human communities
above the most miserable and degenerate in much larger groups; and
the larger those groups, the greater, as a rule, is their cultural and social
development.

How came those larger groups to be formed ? The question may sur-
prise the reader. The formation of a larger group from the original
family group appears to be a quite obvious process. It might take place
in two ways : either a number of neighbouring families might come to-
gether ; or the mature offspring of the original family, instead of separat-
ing from their parents, might remain with them and found secondary
families.

Such a simple view of the origin of human society has been taken for
granted from ancient times until the present . 1 However, it begs the
question, for it assumes that such family groups will continue distinct
within the larger group, and will retain their character and constitution ;
that is to say, it is postulated that those characters are already consoli-
dated in such a manner that they will be preserved unaltered, notwith-
standing the conditions presented by a different type of association.
There is nothing to justify such an assumption. On the contrary, it
follows from all that we know of animal groupings that the association
of a number of separate families must put an end to a form of organiza-
tion which demands isolation as the first condition of its existence.

Whenever, among animals, such an assemblage of a large number of
families takes place even temporarily, the family grouping tends to be
broken up, and the constitution of this larger group lapses into that of
the promiscuous herd. This takes place even among the most typically
pairing and family-forming animals, viz. among birds ; it does so in the
short period in which they gather in large numbers during the breeding
season. The larger grouping no longer depends on the manifestation of
maternal instincts, but on the male instincts. From a maternal group
it is changed into a masculine group, from a family into a herd.

Among the apes we find large assemblages, it is true, and the maternal
instincts are more highly developed than among other animals. But
1 Cf. Westermarck, E.: The History of Human Marriage (1901), p. 49.

THE PRIMITIVE HUMAN GROUP 53

these assemblages are not under the dominance of the males. In human
groups the conditions are quite different. There is no division of labour
amongst the apes ; in the human group, the male is differentiated as a
hunter and a warrior. In a promiscuous human herd a struggle would
inevitably take place for the possession of the females, and would at
once result in the dominance of the most powerful males. The group
would, in fact, be a patriarchal herd, in which a few of the stronger
males would hold the weaker males and all the females in subjection.

In short, unless some principle can be shown to operate which will
maintain the arrangement — unprecedented in biological history — of
an associated group of separate families, the result of simple aggrega-
tion will not be a group of families but a promiscuous herd. We explain
nothing by supposing that a number of families are brought together.

If, on the other hand, the compound group is supposed to be formed
by the offspring of a single family continuing with their parents and
giving rise to new families, the difficulties are even more marked. If
the sons, after they grow up, continue with their parents, they must
either marry their sisters or else import women from another group.
The effect on an incipient human society would be equivalent to the
haphazard fusion of family groups into a herd family. The capture of
females from another group and their introduction into the parental
group would result in a complete transformation of the group’s consti-
tution. If the compound family grouping is to grow out of the original
animal family and to retain its distinctive advantages, the feminine
constitution of the family must be maintained ; it cannot be maintained
if the sons remain and import their wives, while the daughters are taken
away by strangers. The structure of the group would then be trans-
formed. The female line of influence would be destroyed, and a group-
ing, based on male initiative, would be formed. Such a transformation
has occurred in later stages of social evolution ; and, where racial develop-
ment is already fairly advanced, it is not attended with untoward effects ;
but in the initial stages of that development, it would abolish those
conditions upon which the emergence of the social group has depended.
The group would cease to be a maternal animal family; it would become
a masculine herd.

There is, in fact, one way, and one only, whereby the feminine con-
stitution of the family could have been maintained while it expanded
at the same time into a larger assemblage — namely, by the sons leaving
the group, and the daughters remaining and pairing with the males
from some other group.

This peculiar arrangement is, in point of fact, the one which was
adopted by nascent humanity. Everyone who has even a passing
acquaintance with anthropology knows that the most general rule

THE MOTHERS

governing the organization of primitive social groups is that of ‘exo-
gamy’ — the rule that marriages shall not take place within the group but
always with members of another group. The manner in which that
rule is carried out in a large number of primitive societies is by the males
either leaving the parental group and being adopted into the group to
which their wives belong and living with them, or simply visiting them
while they continue to live in their own group. In either case, the
daughters do not leave the family group in which they were born.

The great majority of primitive societies regard the exogamic system
as the most inviolable principle of their social organization. In many
communities it has assumed complicated forms.

The group is subdivided into two, four or even eight sub-groups,
and frequently the members of one group may intermarry with members
of only one other group. With those complexities we need not concern
ourselves here. They are necessary effects of the development of the
exogamic group which in time must needs subdivide. Fundamentally
this principle is identical with the prohibition of incest, for in the primi-
tive family group there is no alternative to the marrying out of the males
except union with their own sisters; thus our own horror of incest
represents the very principle which appears to us so strange in the
exogamic regulations of primitive man.

It is significant that nothing approaching to a satisfying explanation
has been offered of this principle. That failure seems to point to some
radical fault in our method or in our assumptions.

J. F. McLennan, the first expounder of the principle of exogamy, had
no better hypothesis to offer than that exogamy arose from the
scarcity of women resulting from the prevalence of female infanticide
among savages. Both his arguments and his facts have been fully refuted.
Lord Avebury ascribed the law to the practice of capturing wives, but
this would produce results exactly opposite to those that are known to
accompany exogamic organization, and there are, as we shall see, definite
grounds for regarding wife-capture as a much later development than
exogamy. To appeal to the ‘natural horror of incest’ is to beg the question
at issue.

Alleged Injurious Effects of Inbreeding among Animals

It has been thought that the rule of exogamy is connected with the
supposed injurious effects of inbreeding. Though belief in such ill-
effects has been practically universal from time immemorial, attempts
to substantiate it scientifically have failed completely.

Since some writers continue to assume it as a basis for explaining the
constitution of human society, the following points may briefly be
made:

THE PRIMITIVE HUMAN GROUP 55

(i) All animal species propagate without regard to inbreeding, and
some propagate exclusively by what we should term incest. Thus the
African reed-buck usually brings forth two young at a birth, a male and
a female; these, when mature, pair with one another. The same appears
to be true of most of the smaller species of antelopes and all red-deer,
tigers, buffaloes, etc.

(ii) Whole countries have been rapidly overrun by the offspring of
single pairs or of a very small number of individuals. Rabbits have
become a plague in Australia ; they are the progeny of a few individuals
brought there in 1863. In New Zealand a hardy race of wild pigs, much
sturdier than any of our domesticated breeds, is the offspring of a
couples of sows and a boar left there by Captain Cook on his second
voyage, and of a few animals similarly turned loose in subsequent years.

(iii) Numerous experiments have been carried out on rats, mice,
guinea-pigs and rabbits, by causing them to inbreed closely for many
generations. They have in most instances shown no perceptible evil
effects as regards the quality and size of the animals, but a diminution
in fertility has been observed. It is, however, well known that all ani-
mals, especially rodents, suffer a diminution of fertility when kept in
close confinement. In recent large-scale experiments by Dr H. D. King
on white rats, it was found that: ‘These laboratory rats, which have
been inbred as closely as possible for twenty-two generations, are in
every respect superior to the stock rats from which they took their
start six years ago, and which have since been bred in the usual indis-
criminate manner.’ The largest albino rat ever recorded was produced.
The inbred rats ‘live fully as long as do the stock rats, and they appear
equally resistant to disease’. The fertility of the inbred rats was nearly
8 per cent greater than that of the stock rats.

(iv) The experience of breeders of domesticated animals has been
supposed to afford evidence that inbreeding results in degenerative
changes and impaired vitality and fertility. The English racehorse is
one of the most closely inbred of existing animals. Several elaborate
investigations have been made of the stud records of famous English
racehorses, and it has invariably been found that noted winners excel-
led in almost exact proportion to the close inbreeding shown by their
pedigrees. The Percheron horses are also a closely inbred race, most of
the individuals being traceable to two original sires. Numerous similar
illustrations of exceptionally fine breeds of domestic animals, the pro-
ducts of prolonged inbreeding, are mentioned by Darwin in his book on
The Variation of Animals and plants under Domestication — a work in
which he seeks to uphold the evils of inbreeding while showing the
benefits sometimes observed in domesticated animals from occasionally
crossing them with another breed.

THE MOTHERS

Inbreeding may indeed, produce injurious results where the animals
are bred for the improvement of a single characteristic: thus pigs bred
for fat sometimes become less fertile and the sows are unable to nourish
their young. The surprising circumstance is that such evil effects are
not more common and more conspicuous. It is only where the stock
has been thus inbred that the ‘occasional cross’ is beneficial. In Central
Europe and the Baltic countries the art of stock-breeding was for long a
complete failure; until recently nobody ever heard of a reputable breed
of German or Scandinavian horses or cattle, and breeds of domesticated
animals had constantly to be improved by importations from England,
France and Spain. The chief reason for this was the great authority
attached in those countries to the works of Professor Settegast, who
fiercely denounced inbreeding in any form, until Count Lehndorff and
de Chapeaurouge exposed the baselessness of his dogmas.

If the results of the breeding of domesticated animals prove anything,
it is the absolute innocuousness, if not the actually beneficial effects, of
inbreeding. Thus in many British parks, such as Cadzow Castle,
Chartley, Lyme Park, Somerford, herds of cattle have been left to in-
breed from time immemorial. The Arabs inbreed their famous horses
as closely as possible without any scientific refinements. In the records
of breeding from domesticated animals there is not a single fact, alleged
or verified, which indicates, much less evidences, that inbreeding, even
the closest, is in itself productive of evil effects.

Results of Inbreeding in Human Communities

In the human race the evidence is, if anything, even more definite.
Close inbreeding takes place habitually among many peoples ; for, even
where the principle of exogamy is strictly observed, if marriage outside
the group takes place for generation after generation in one particular
other group, the intermarrying members stand to one another in the
relation of cousins. It is a very widespread custom for a man to marry his
first cousin; such marriages are regarded in many parts of the world as
obligatory. Yet nowhere have any evil effects been observed, and the
races which practise those marriages include some of the finest physical
types of mankind. Thus the Bataks of Sumatra are described as being
physically the best-developed race in the Indian Archipelago; the men,
Junghuhn remarks, might have stood as models for the sculptors of
Greece. An elaborate census of the population of Fiji, among whom
marriage between first cousins was regarded as a sacred duty, showed
that such marriages were associated with a higher birth-rate and a
markedly greater vitality in the offspring than unions between non-
relatives. So much so that the former are maintaining their numbers
while the latter are rapidly dying out.

THE PRIMITIVE HUMAN GROUP 57

From the large number of peoples who favour cousin-marriages no
instance has been brought forward which might be set against the testi-
mony of such instances. 3 In almost every part of the world small, iso-
lated communities exist where, for centuries, marriages have of necessity
taken place between closely related individuals. They are almost invari-
ably distinguished by conspicuously fine bodily development and robust
health. For example, in the Tengger Hills of Java the Surabaya com-
munity, numbering some 1,200 people, has intermarried for ages. ‘They
differ from the people of the plain, being taller and more robust*. The
same is true of a small segregated community, the Baduwis, in West
Java, who number no more than about forty familes all told. In Europe
such intermarrying communities are common enough in hill districts
and among fisherfolk. That of Batz, near Croisic, which numbers about
3,300 people, was made the subject of a very thorough investigation by
Dr Voisin. He did not find an instance of malformation, mental disease
or any of the evils ascribed to inbreeding; marriages between first
cousins in that community were found to produce an average of 4-6 off-
spring, whereas the general average for France at that time was only
33. The island of St Kilda contains [i.e. in 1927] twenty-seven families,
and supplies, as Dr Kerr Love remarks, the proper conditions for the
production of all the evils ascribed to intermarriage. The inhabitants
are poor and badly housed. Dr C. R. Macdonald, the Chief Medical
Officer of Health for Ayrshire, has assured the writer ‘that there is no
history of any case of deaf-mutism in this remote islet, nor, moreover,
of other signs usually attributed to the results of intermarrying.* In an
article (1888) he says: ‘There are no cases of deaf-mutism; insanity and
idiocy are unknown, and cases of imbecility are extremely rare.* This is
after centuries of intermarrying.

Many similar examples are forthcoming. The Pitcairn islanders are
descended from nine of the mutineers of the Bounty who in 1790 were
deposited on the island, together with six men and twelve women from
Tahiti, and in 1800 consisted of twenty-five people. They are described
by all observers as remarkably strong, healthy and well-built; the men
averaging six feet in height, both sexes being well formed and hand-
some; the women being as muscular as the men, and taller than the
generality of women, and the children uniformly healthy.

While the manifestations of racial degeneration in royal families, such
as the Hapsburgs and the Spanish Bourbons, are often popularly
referred to — it is hard to see on what grounds — as illustrating the evils
of inbreeding, those royal families which have systematically practised

1 Dr Westermarck has ransacked ethnological literature for examples, yet he
feels compelled to apologize for the results and for ‘their vagueness and more or
less hypothetical character*.

THE MOTHERS

dynastic incest furnish no evidence of those supposed evils. None is
afforded by the Ptolemies ; the practice which they adopted when they
took over the throne of Egypt had been regularly observed in that
country for at least 3,000 years. In the golden age of the Egyptian
monarchy, during the XIXth and XXth Dynasties, every king of the
former married his sister as the lawful mother of the heir to the throne.
Yet the race that produced Seti and Rameses affords no evidence of
degeneration, nor does there exist in the age-long records ofbyfarthe
longest line of kings in the world’s history, among whom, not mere
inbreeding but actual incest was a fundamental principle, any fact
lending support to the doctrine of the evil results of inbreeding.

No attempt to demonstrate this theory by statistical investigation has
met with any success. The most thorough general investigation of the
kind is still that undertaken by Sir George H. Darwin, who was, like
his father, strongly disposed to believe that inbreeding is attended with
injurious effects. As regards fertility, he found that balance was slightly
in favour of cousin-marriages. He found no differences as regards
insanity or deaf-mutism.

When the belief that consanguineous marriages produce deaf-mutism
and other physical conditions, such as mental deficiency and sterility,
are scientifically examined, they turn out to be equally unjustifiable. Thus
a Mr Huth, at the end of an exhaustive discussion, declares that ‘statis-
tics on which so much reliance has been placed as a proof of the harmful-
ness of consanguineous marriages are, when not absolutely false, miser-
ably misleading and defective’.

The belief that the offspring of parents consanguineously related will
be stricken with deformities or diseases is much older than any attempt
at scientific inquiry, and in more recent times is often derived from the
fact that such marriages are condemned by the Roman Catholic Church.
But this belief is firmly held not only in Europe but among almost all
savages. The mountaineers of Albania believe that incest results in
deaf-mutism; the Aleuts that it produces malformations and that the
offspring of such unions will have tusks like a walrus. The Kaffirs
believe that it causes idiocy, and the Basoga of Uganda are even scan-
dalized at the thought of incest being committed by their cattle and
punish the criminals. The consequences are not limited to the children,
but may extend to the parents and even to the more distant relatives or
sometimes to the whole tribe. In the Celebes such unions are regarded as
causing earthquakes and floods and volcanic eruptions. (These are, of
course, the usual effects of infringing a tabu.)

But the definition of what constitutes a consanguineous marriage
varies widely. Thus the Murung of Borneo permit marriage between
brothers and sisters, and believe that their offspring will be remarkably

THE PRIMITIVE HUMAN GROUP 59

healthy, yet regard with horror marriages between cousins and certain
other relationships. The Bhotias prohibit cousin-marriages on the
father’s side only; while the Herero consider that, while the children of
a brother and a sister may marry, the children of two brothers or two
sisters may not. Some people regard all marriages between close rela-
tives as particularly lucky. The Kalangs of Java believe that incestuous
unions between mother and son are blessed with prosperity, while the
tribes of British Central Africa believe that incest makes one bullet-
proof. Current ideas of the evils of inbreeding are, in fact, a survival of
an ancient superstition.

Theories of the Origin of Exogamy

Sir Henry Maine and Lewis Morgan accounted for incest prohibi-
tions and exogamous practices by supposing that men noticed the evil
results of inbreeding and forbade it. Many savages give a similar account.
Yet it may be noted that they give no similar account of marriage be-
tween extremely young persons which invariably produces undersized
offspring of poor vitality. The physical decay of many primitive peoples
is due to this practice, but it has scarcely anywhere led to a prejudice,
still less a rule against it. Although the belief that inbreeding is damag-
ing has now been generally abandoned, Dr Westermarck, while profess-
ing his inability to find conclusive proof, nevertheless decides that
‘inbreeding generally is in some way or other more or less detrimental’,
and suggests that the peoples who failed to preserve such principles
were exterminated by these injurious effects. The practice of exogamy
itself he ascribes to the lesser sexual attraction of habitual companions.

M. Durkheim starts from the proposition that defloration of a female
of one’s own group must be regarded with horror. Unfortunately
defloration in primitive society does not happen in marriage, but is
usually performed by the very members of the group whose relations
with the females would normally be considered incestuous . 1

Darwin believed that the emigration of young males was due to the
jealousy of the old males, but such a procedure — even if it exists — could
not account for the abhorrence of incest; and the tyranny of the older
males would surely operate as strongly against strange males as against
their own male offspring, so that young males would thus never obtain
wives in a strange group, except by fighting old males.

Havelock Ellis is one of those who feel that the explanation is ‘really
exceedingly simple’ — boys and girls brought up together from infancy
fail to evoke the pairing impulse because familiarity has dulled the
sexual stimulus. But, as is widely known, men frequently marry the
companions of their childhood, and relationships which have for a long

1 See p. 397.

THE MOTHERS

time been platonic end in marriage. Dr Ellis confounds the sexual
impulse with the mating impulse. The latter is based on companionship
and affection and not on sexual desire; indeed, it involves subordination
of the male sexual impulse.

In primitive societies in every part of the world children are
commonly betrothed to one another and grow up like brothers and
sisters. Such marriages are attended with the greatest success. Thus we
are told of the tribes of the Upper Congo that it is rare for children
brought up together to fail to marry and to dislike one another. Drs Hose
and McDougall, speaking of the intercourse of a youth and his sister by
adoption in Borneo, point out the difficulty of reconciling these facts
with Westermarck’s theory. The occurrence of brother-sister incest and
the strong objection of the Sea Dayaks to newphew-aunt incest — these
often being members of distant communities — seem fatal, they add, to
Westermarck’s theory.

The fact, adduced by Mr Walter Heape, that animals will neglect
their accustomed harem to attend a newcomer does not illustrate the
greater attraction of the stranger but the male’s instinct to impregnate
as many females as possible. And he does not explain why the buck
rabbit, which is so susceptible to the fascination of the strange female,
is yet incorrigibly endogamous.

Derivation of the Rule of Exogamy from the Constitution of the Maternal

Group

All these theories have one feature in common — they refer exclusively
to the operation of the male sexual instincts and assume a patriarchal
social state in which those instincts are dominant. As soon as we postu-
late that primitive human groups were not patriarchal but matriarchal,
we see that observance of the rule of exogamy is essential to the pre-
servation of the maternal character of the group. If the women left their
family to join their husbands, that family would cease to be a maternal
group. If the men were the sexual mates as well as the brothers of the
women, patriarchal succession would be established, and their authority
and rivalry would bring about patriarchal dominance. The mothers are
the basis and bond of the primitive group, and the only relation origin-
ally taken into account is the maternal relation. Kinship and descent are
reckoned exclusively through the women; the relationship through the
father is ignored. To permit women to follow strange men, to sever their
connection with the group, would be to break up the social unit, and
would be opposed to its most basic sentiments.

Such principles do not, of themselves, impose upon the men the need
to mate outside the group, but the distinction between the permanence
of the women and the freedom of the men being established, there are

THE PRIMITIVE HUMAN GROUP 6l

many factors to lead men to make use of their freedom. Thus while men
are naturally of a roving disposition, the women are home-loving. The
man instinctively seeks for food, the woman to settle down. But while
such facts make for the association of males with females from another
group, they do not imply an actual prohibition of sexual intercourse
between members of the same group. However, a habit which has
hardened into an established rule that one should seek one’s partner
elsewhere inevitably gives rise to a corresponding prohibition, not to
seek them in one’s own group. This prohibition is reinforced by the
sense of ownership implicit in maternal instinct. In the human family
the maternal instinct does not, as in animals, cease to operate when the
males reach sexual maturity. It is not the mother, therefore, who drives
off her male offspring, but the son who at sexual maturity tends to
transfer his allegiance to another female. The animal is wholly unreason-
able and responds at once to impulse ; the semi-human mother was not
the gentle creature of our conceptions, but a wild animal to whom the
idea that one of her sons should transfer his allegiance from herself to
one of his sisters was a horrible thought, because a blow to her jealous
love. Darwin describes how female baboons protect their young from
being teased by one another; since any attempt by a male to perform
intercourse with another member of the brood would be regarded by
the female as violence, this would infallibly bring the mother to her
assistance.

Though it is not a trait likely to be noticed much by observers, there
are several references to the jealousy of primitive mothers. In New
Britain ‘sometimes fond old mothers are desirous of keeping their sons
with them as long as possible’. One very intelligent Dutch peasant said
that mothers are more jealous with regard to their favourite son than
wives as regards their husbands ... ‘if mothers had their own way their
sons would never marry — at least not for a long time’.

The incest prohibition applies primarily to relations between
brothers and sisters. In the simplest forms of exogamy, as found in
South-eastern Australia, where there are two intermarrying classes, the
system prevents marriage between brothers and sisters but not between
parents and children. In the Island of Kiwai, a father may take his
daughter to wife, although brother-and-sister unions are regarded with
abhorrence. Similarly in some of the Solomon Islands. It is easy to see
why this should be extended from the sister to the mother, for the awe
and dread attaching to the head of the family would make it even less
likely that she would be the object of incestuous advances.

These rules are, of course, generalizations. It is most unlikely that in
every primitive human group the males sought their partners elsewhere,
and instances in which incestuous relations habitually take place have

THE MOTHERS

already been noted. And if this prohibition is the most primitive of all
the regulations imposed upon sexual relations, every subsequent regula-
tion must have tended to consolidate it.

The Mother-in-law

Connected with the rule of exogamy is the fact that in savage society it
is a constant rule that a man may not speak to his mother-in-law and
generally may not even look at her; the breach of this rule is regarded
with as much horror as incest. In Australia a man is warned of the
approach of his mother-in-law by the sound of a bull-roarer; it was
formerly death for a man to speak to her — later banishment. In Tas-
mania a native, concerned about the attentions a younger man was
paying to his wife, betrothed his new-born daughter to the suspected
rival — thus making it quite impossible for the latter even to look at his
future mother-in-law. In New Britain a man must not only not look at
his mother-in-law, but must avoid meeting her and must hide if he sees
her coming. Suicide of one or both would be the only course if he
inadvertently spoke to her. In the Banks Islands a man would not follow
his mother-in-law along the beach until the tide had washed out her
footsteps. The rule is as rigorous in Africa as in Australia and Melanesia.
Among the Baholoholo the ceremonial avoidance is observed even after
the death of the wife. The Yukatans believed that if a man even met his
mother-in-law he could never beget children. The Baganda regard the
prohibition of meeting the mother-in-law as more sacred than the
prohibition against incest, and the remark would appear to be generally
applicable. The well-worn jokes of the British music-hall are probably
therefore an echo of a very primitive sentiment for which no satisfactory
interpretation has ever been offered.

Lord Avebury’s hypothesis that the rules of mother-in-law avoidance
are connected with wife-capture is ruled out by the fact that those usages
are commonest where wife-capture does not obtain, and are rare where
it does. Moreover, they would not account for singling out the wife’s
mother for avoidance, but would require rather avoiding the menfolk
of the wife’s tribe, and especially brothers and uncles on whom the
duties of revenge would fall. The rule of avoidance does in several
instances extend to other relatives, and even in Australia to the whole
clan, but in most cases it applies to the mother-in-law alone and where
others are included to her chiefly. Father-in-law avoidance by women
is found f6r the most part in advanced Oriental nations, such as the
Hindus and Chinese, where patriarchal domination is established.
Significantly, avoidance rules also sometimes apply to the grandmother.
The suggestion frequently made that these rules are intended to prevent
improper intercourse between the mother-in-law and the sons seems

THE PRIMITIVE HUMAN GROUP

to be excluded by certain curious derivative practices found in several
tribes. Thus among the Navaho, although mother-in-law taboos are
exceptionally strict and extensive, yet a Navaho can sometimes avoid
them by the simple expedient of marrying his mother-in-law pro forma
before he marries the daughter. The Cherokecs and the Caribs have hit
on the same plan, while with the Wagogo and the Wahele of East
Africa a man must cohabit with his future mother-in-law before he is
allowed to marry the daughter. It seems clear that tabus which can be
evaded by marrying one’s mother-in-law are not intended to guard
against incestuous relations with her.

On the view that the feeling against incest and the rule of exogamy
are consequences of the matriarchal character of primitive groups,
these avoidance customs present no problem. It is because he fears the
mother-in-law that primitive man originally avoided her so sedulously.
It is a memory of the awe which was originally inspired by the mother
that, even where the supremacy of the mother has passed away, it still
survives in these rules whose widespread distribution betrays their
original importance.

Moreover, mother-in-law restrictions can be commuted by presents.
Among the Arapahos, all restrictions are removed if a man presents his
mother-in-law with a horse. That the supposed danger which these rules
were designed to obviate comes from the mother-in-law and not from
the son is further shown by the fact that among some peoples (e.g. the
Warramunga of Central Australia), though a man may not go to a camp
where his mother-in-law resides, she is quite free to visit him in his own
camp. The idea underlying these observances is also betrayed in other
instances. Among the Banyoro of East Africa, for example, a man is
not obliged to avoid meeting his mother-in-law, but it is absolutely
essential that he kneel down and remain in a reverential position when-
ever he meets her. In these and other customs the mother-in-law does
not appear as a possible object of unlawful desire, but as an offended
personage who it is needful to conciliate.

If my conclusions concerning the origin of the rule of exogamy are
correct, such customs present no real enigma, but confirm these con-
clusions and serve to exhibit the natural authority of which the primi-
tive group was the expression.

CHAPTER 6

The Motherhood

Montesquieu was considerably amused when he read in a mission-
ary’s account that in Formosa the groom remains in the house of the
bride’s parents, but in fact this custom, which is known as matrilocal
marriage, is widespread, and is of especial interest since it suggests the
existence of former female supremacy.

It is found among many varieties of Eskimo; it was formerly the
general rule among all North American Indians. The Senecas, the most
important of the members of the Iroquois confederation, formerly
dwelt in ‘long houses’ under the authority of a matron. The female
portion ruled the house, and any man who disobeyed orders might be
forced to retreat to his own clan or to leave and start a new matrimonial
alliance elsewhere. Among the Crees, when a man marries he resides
with his wife’s parents, who treat him as a stranger till the birth of his
first child.

Matrilocy is also a feature of the Pueblo Indians of New Mexico. It
has been suggested that, as these are the most advanced in culture of
North American Indians, matriarchal society is not primitive but a
product of advanced development. However, the matriarchal organiza-
tion is also found in an even more absolute form in the rudest and most
uncultured tribe in the whole North American race, namely the Seri
of the Californian gulf. To say that they are in the Stone Age is scarcely
accurate, for they do not even fashion stones but merely pick up a cobble
for crushing bones or severing sinews. When provided with knives,
they do not know how to use them. No other human tribe is so devoid
of material devices. They have no form of agriculture, they do not cook
their food, and, in eating a heavy joint, so rotten that the flesh could
easily be scraped off with a knife, they lift the whole thing to the mouth
and gnaw it. So fierce is their hostility that it has not been possible to
observe them as fully as is desirable ; they have murdered many would-be
observers.

The most noticeable feature of their organization is the prominence
of the females. The social unit is the maternal clan, determined by
descent from a common line of mothers. The clan is headed by a clan-
mother and comprises a hierarchy of daughters and granddaughters.

THE MOTHERHOOD

The indigenous name of the tribe is ‘kunkak’, which means Womanhood
or Motherhood. Their rude dwellings are erected by the women with-
out help from the men, and belong exclusively to the matrons. It was
often difficult to identify the husband, partly because he was often in-
congruously younger than the mistress, and partly because of his lack
of authority. The females have no term for father, and there is some
question as to whether the Seri recognize paternity. The women are the
only real workers ; the men tacitly accept the decisions of the mother.
Though male chiefs are elected, mainly for leadership in war, all magi-
cal powers are considered to reside in the women, and the election of a
chief depends largely on the magical powers of his wife. A man generally
marries all the sisters of one family and there are indications that for-
merly, when the men were numerous, all the brothers of a family were
conjugally bound to all the sisters of another. The prospective bride-
groom is subjected to a series of elaborate tests before acceptance by
the mothers.

Matrilocal marriage is also widespread among the Caribbean races,
and was practised in Mexico by the tribes of Yucatan, as also in Peru
under the Inca monarchy. Old missionaries noted it with astonishment
among the Mozcos of New Granada and among various tribes of the
Orinoco. It is reported of the Gran Chaco tribes, of several tribes of
Southern Argentina and those of Tierra del Fuego.

In Africa, the matrilocal rule is strictly observed both among such
primitive tribes as the now almost extinct Bushmen, the Basuto, the
Barolong and all other Bechuana tribes. The practice is very general in
East Africa, chiefly among the Marotse, the Yahos, the Anyanga, the
Tumbuka, the Wakamba, the Mosuto, as well as among the primitive
pygmies of the Congo forest.

The social constitution appears to have been characteristically matri-
archal throughout the vast region which extends south of the Sahara,
from the Atlantic to the Nile, property being transmitted by a man to
the children of his sister, and matrilocal marriage being general ; thus
among the Kona, the Fulani, the Kilba, the Kulangas, the Madi, the
Buduma and Baele of Lake Chad, the Nuer of the Upper Nile, the
Barabra of Nubia and many others. Special interest attaches to the white
races of Northern Africa who now inhabit the Sahara region, for, accord-
ing to a common view, these populations are direct representatives of
the race which laid the foundations of Western civilisation on the islands
and shores of the Mediterranean. These people are now matriarchal and
there are strong grounds for thinking that these early invaders were also
pronouncedly matriarchal in character. All the peoples of Northern
Africa west of Egypt belong to the Berber race (which the Greeks
called Libyans), a white race scarcely different from the inhabitants of
t.m. — 3

THE MOTHERS

Southern Europe, some being so fair they might pass for Irishmen or
Scotsmen. The Berbers of Algeria and Tunisia are now largely Muslim
and thoroughly patriarchal, but those which withdrew to the interior
rather than yield to a foreign invasion — the Tuareg — have preserved
both their ancient language and their matriarchal constitution. Descent
among them is reckoned in the female line, and property descends
from a man to his sister’s children. Though there is an Eve in their myth
there is no Adam. Their ancestors in Roman times, the Numidians,
had the same customs, and were named after their mothers. Ibn Batuta
was the first modern traveller to describe them. He says: ‘The women
... are not timid in the presence of men, nor do they cover their faces
with the veil. . . . The women do not follow their husbands, and should
any of them wish to do so, her relatives would prevent it.’

Every traveller has commented on the independence of the women
among the Berber races of the Sahara, such as the Tibbu. ‘In all matters
their word is law.’ The culture of the Tuareg is almost exclusively
confined to the women (the men are entirely illiterate); and it is they
alone who preserved knowledge of the ancient Libyan tongue and
script, which appears to resemble that of Minoan Crete.

The Malay race which has spread over the whole Indonesian region,
sending off-shoots as far as Madagascar and China, has for centuries
been under the influence of Hindu and Islamic religions, but they always
distinguish between the customs of their adopted religion and their
traditional law known as adat , and cling persistently to the latter. Many
Malays prefer the older form of matrilocal marriage to the new patrilocal
form introduced by Islam. Chinese travellers of the time of the Ming
Dynasty (1368-1643) noted the Malays of Sumatra were matrilocal.
Pistorius describes how the Malays of Negri Sembilan live in barrack-like
rows of dwellings, forming a Sa-mandei or Motherhood, consisting of a
house-mother and her descendants in the female line. He was much
amused to see the husbands walking across at dusk from their own homes
to join the wives; if a man has several wives, he visits each in turn.
(This practice is also true of many other matrilocal tribes where polygamy
is practised.) The husband has no obligation to maintain his wife; this
falls on the maternal family, and the head of the family is usually the
brother of the mother. The property of a Malay passes on his death,
not to his wife and her children, but to his brothers and sisters and to
his sisters’ children; that is, to his maternal family.

An unmodified matriarchal organization is found in Northern Su-
matra, among the Orang Mamaq, who are divided into strictly exogamous
clans : husband and wife seldom live together, but if they do the husband
comes to his wife’s clan. The Sakai tribal organization is strictly matri-
archal, and it is the rule among the primitive races of Timor, of the

THE MOTHERHOOD

neighbouring islands of the Southern Malaccas and in the Celebes. In
the Northern Malaccas, as in Java, Islamic marriage customs have now
become universal, but in many of the islands between the Celebes and
the Philippines matrilocy remains the practice. It is also the rule in
Borneo among both land and sea Dayaks, and was formerly the general
native custom in the Philippines. The Spanish conquerors noted that
the Pintados took sides with their wives’ relations, even against their
own fathers and mothers, and thought that this indicated that they loved
their wives very dearly.

The natives of the Micronesian region — the Carolines (with the
exception of Yap), the Marshall, Mortlock, Pelew and Gilbert Islands —
are matriarchal in their social organization. Similarly in the western
islands of the Torres Straits, where matrilocal marriage is the rule. It
also appears to be the typical usage in most parts of New Guinea,
though there are many gaps in our information about that vast island.
Several regions appear to be in a state of transition from matrilocal to
patrilocal usages. Thus in some parts of Dutch New Guinea, a man
may take his wife to his home for a year, after which she returns to hers
and he visits her. The fundamental matrilocal character of their cus-
toms is indicated by the fact that no boy can go through puberty
ceremonies until he has resided for a while with his mother’s family.

Evidence is also adduced from New Zealand, from the Marquesas,
the Nicobar Islands, the Ainu of Japan, the Kurile Islands, etc. In
Northern and Central Asia, the custom that the bridegroom resides
for a more or less prolonged period with his wife’s family or that the
bride, after a short residence with her husband, returns home for a
prolonged period, is widespread. These customs, so similar to the
practices in New Guinea and Africa, where we know that the culture is
in a state of transition from matrilocal to patrilocal usages, suggest that
this is the case in Asia also ; an inference which is confirmed when we
find, in several cases, that this is indeed occurring. Thus among the
Buryat, a Mongolic tribe of Southern Siberia, the bride returns home
after marriage for six months or more, and this visit is several times
repeated. Buryat tradition states definitely that it was formerly the
usage for the husband to take up his abode permanently in the home of
his wife.

Matriarchal societies are found in several parts of China, among
aboriginal populations of non-Chinese race. Thus the Nue’kun is said
to be permanently ruled by a woman, and the tribes in the mountains
of Kwei-Chow are matrilocal. Matrilocal marriage is characteristic
among the peoples of Siam, Burma, Indo-China and Tonkin.

In India, in the hills of Assam, there are various tribes so little dis-
turbed by change that they still erect large standing stones like the

THE MOTHERS

menhirs of Brittany over their dead. The social unit of the Khasi tribe
is called Mahari, that is ‘Motherhood’, and Sir Charles Lyall says it
‘presents one of the most perfect examples still surviving of matriarchal
institutions, carried out in a logical and thorough manner which, to those
accustomed to regard the status and authority of the father as the foun-
dation of society, are exceedingly remarkable’. The mother is the only
owner of real property. Among the Synteg, the husbands only visit
their wives at night and are known as ‘the Begetters’. A similar organiza-
tion is found among other tribes of the region, such as the Garos.

Among other of the aboriginal races of India, where Hindu customs
have not influenced the original institutions, matrilocal marriage is
found, as among the Kehal, a tribe of boatmen on the Indus, who are
now Mohammedans, among the Gonds, the Santals and the Mundas.

The term ‘the Motherhood’ is also applied to the family group
among the famous Nayars of Southern India, where the household is
constituted by the mother and her children, and a Nayar husband may
not even partake of food in his wife’s home, but must visit after supper.
At the present day much of the social organization of these Motherhoods
has disintegrated, but the matrilocal element still remains strong.

In several of the above instances, marriage is not permanently
matrilocal, but the woman resides in her own home for a limited period
after marriage. Sometimes this becomes a mere ceremonial, a period
of a few days or even simply the wedding night. Perhaps the most
attentuated form is found among some tribes of Southern India, such
as the Mappellas, where the bride and bridegroom are locked up for a
few moments in a room in the bride’s home after the wedding cere-
mony. The marriage is supposed to be consummated during these
moments, though, in fact, the custom is purely ritual. In our own society,
the practice survives in the practice of taking the wedding-lunch at the
bride’s house.

In many parts of Africa and among uncultured societies elsewhere,
even where the wife is brought to her husband’s home, connection with
her own family remains much closer than is the rule in advanced
patriarchal societies. Missionaries complain of the way the women tend
to go back to their mother on the least occasion.

Such limited matrilocy is sometimes difficult to distinguish from a
marriage by service in which the bridegroom gives his services for a
stipulated period, and it is sometimes thought that these practices
merely represent a commutation of a payment. We shall consider the
matter more fully later, and it will become clear that, on the contrary,
making a payment is a commutation of service rather than the reverse.
Marriage by service is found nowhere except where permanent matri-
local marriage is also customary or is known to have been formerly

THE MOTHERHOOD

general. Thus among the tribes of Assam* where complete matriarchal
organization obtains, we are told that among the Bodo and Dhimal, who
now pay a ‘bride-price’ on marriage* ‘a youth who has no means of dis-
charging this sum* must go to the home of his father-in-law-elect and
literally earn his wife by the sweat of his brow, labouring “more
judaico”, for up to seven years.’ Indeed* in some instances, instead of
the service being regarded as a form of payment, the husband himself is
paid by the wife’s parents in order that he should forgo any claim to
remove her and her children from their home.

Missionaries, with the Biblical precedent of Jacob’s marriage in their
mind, have frequently misreported matrilocal marriage as ‘serving for a
bride’. But in many of their accounts it is clear that this is a misunder-
standing based on the fact that, after the end of the service period,
frequently on the birth of a child, the relationship is terminated, not
confirmed. In some cases, the fact that the husband erects his own wig-
wam, because of the limited accommodation in the wigwam of the
wife’s parents, has been taken to be the termination of his residence with
the family, but it is clear that this is not a taking-back of the woman to
the husband’s clan. Indeed, the marriage of Jacob which causes this
misunderstanding is itself not an example of the commutation of services
by a payment, for the Bible expressly tells us that Jacob’s father-in-law
denied that he had any right to remove his wife even after twenty years
(Genesis xxxi. 26, 43).

As we have seen, the fact that the female, not the male, determines the
dwelling-place arises from the biological fact that it is the female of the
species who chooses a suitable lair. (All animals may be said, in so far as
they form sexual associations, to be matrilocal in habit, and it is natural
to infer that habits of primitive humanity were similar.) The validity of
this inference is proved by a social fact to which there are no exceptions.
Whenever the man removes the wife from her home and brings her to
his own, the procedure invariably involves a transaction whereby the
woman’s family sanctions this removal. In all but the highest cultures
this sanction involves the payment of a compensation. 1 Among the
Alfurs of Ceram, a man has the option of marrying his wife without
payment and taking up residence in her village, or of paying a bride-
price and removing her; but if he marries a woman of the same village,
there is no question of payment. The same is true among the Alfurs of
Buru and in the Kei Islands. Hence there is no alternative but to con-
clude that the practice of matrilocal marriage was the original form of
marriage union and is coeval with the origin of humanity.

1 A woman might, it is true, be removed from home by abduction, but, as we
shall see, it is impossible that this was ever the usual and general mode of
obtaining a wife.

THE MOTHERS

This conclusion is confirmed by ethnological documents, for though
there are numerous societies where matrilocy clearly prevailed at a
former date, there is no single case where it appears that patrilocy
preceded matrilocy.

The usage most frequently associated with matrilocy is matriliny, or
the reckoning of descent from the mother and not from the father.
When Herodotus noted it among the Lykians in Asia Minor, he thought
it singular, but we now know it to be the rule with about half the people
of the world below the most highly developed stages of culture. With
most of those who reckon descent in the paternal line, clear evidence
exists that the opposite was formerly the case.

It is sometimes thought that matriliny is the most distinctive feature
of matriarchal society, but, in point of fact, matriliny is compatible with
a very depressed status for women, as for example, among the majority
of Australian tribes. On the other hand, matrilocy inevitably gives a
woman a high status, for she remains among her blood relations while
the husband is more or less a stranger. As Dr Kroeber says of the Zuni :
‘Upon her permanent occupancy of her house rests the matrilinear
custom of the tribe.’ The husband is economically destitute, for all the
property is owned by the wife. Matrilocy necessarily implies a matri-
archal form of society.

The Status of Women in Uncultured Societies

It used to be asserted that the more civilized a society the higher the
status of the women in it. It was held that in uncivilized societies their
position was one of outrageous oppression. If this were true, the Red-
skins and the Papuan cannibals would have to be accounted more civi-
lized than the Chinese and the ancient Greeks. In point of fact, the exact
reverse is the case. In the great majority of uncultured societies, women
enjoy a position of independence and of equality with the men, and
exercise an influence which would appear startling even in the most
feministic of modern societies.

It is true that there are a few primitive societies where women are
subjected to the despotism of brute force. Thus among the Australian
aborigines the condition of the women is utterly degraded, and they are
treated with ‘about the same consideration as the dogs’. A girl of seven
to ten is handed over to a man old enough to be her grandfather. For
the slightest offence she is beaten with a yam-stick and not infrequently
speared. Government records in Adelaide furnish numerous cases of
Blacks murdering their Lubras. In times of scarcity, she is the last to be
fed. ‘Few women’, says Eyre, ‘will be found upon examination to be
free from frightful scars upon the head or the marks of spear wounds
about the body.’ ‘It is not uncommon’, says Mr Hodgson, ‘to meet a

THE MOTHERHOOD

7 *

woman with scarcely a single hair to be seen, from the frequent strokes
which have descended upon her unfortunate pericranium.’ Any female,
whether old or young, if found unprotected, is invariably ravished and,
in most instances, killed afterwards. Queensland natives chastise their
wives by rubbing hot coals over their stomach.

A similar state of things is found in most Melanesian islands. ‘The
New Caledonians take no more account of a woman than of a pig.’ A
New Caledonian chief, having acquired an old flintlock, practised
shooting, using women set up in rows as targets. Another native, who
was converted by a missionary, asked to be received into the Church,
but was told that he could not be admitted as he had two wives. He
returned next morning saying that all was in order, as he had now eaten
the superfluous wife.

In some parts of Africa things arc as bad. The Bangala of the Congo
quite commonly eat their wives. In 1887, a Bangala chief informed a
missionary that he had eaten seven of his wives, and had invited their
relatives to the feast, in order that there should be no family unpleasant-
ness. Among the Somalis, as among the Sifan of Tibet, the bride begins
married life by receiving a sound flogging from her husband.

In low phases of culture where man is dominant such a state of affairs
tends to come about. No innate scruples and no external compelling
force restrain primitive man from using his power ruthlessly towards the
women whom he regards as his property.

Yet there arc numerous exceptions universally distributed; a few have
already been noted. Among the Eskimo, ‘the women appear to stand
upon a footing of perfect equality with the men’. Among the North
American tribes, the position of the women is one of complete indepen-
dence. As the missionary Lafitau said: ‘Nothing is more real than the
superiority of the women.’ The importance of the counsels of the women
among the Iroquois confederation is shown by the fact that the deeds of
land-transfer of the colonial government nearly all bear the signatures of
women. The compensation due for the murder of a woman was double
that for the murder of a man. Similar data may be adduced for Pre-
Columbian Central America, Western Peru, New Britain and many
parts of Melanesia. In Dutch New Guinea, a traveller saw a man sub-
jected to a sound drubbing from his wife because he had brought some
trinket to the boats for barter. Throughout the Malay archipelago
women are treated with uniform consideration, and in all parts of
Micronesia the position of women is notably exalted. As one missionary
reported: ‘The women in this country have arrogated to themselves
those rights which everywhere else are claimed by the husband. The
wife absolutely rules the house: she is the master and the husband is
unable to dispose of anything without her consent.’ If he displeases her,

THE MOTHERS

she maltreats him or quits him altogether. Throughout Polynesia, the
position of women, though theoretically subordinate, is invariably one
of great independence and influence. Similar data can be adduced from
many African tribes, including the Bagisu, the Ekoi, the Bega, the
Bantus and many others. It has already been noted that similar condi-
tions obtain among the Tuareg.

In Asia, while among the most highly civilized races, such as the
Hindus and the Chinese, women occupy a position of great subordina-
tion, among the most primitive and secluded races their status is high.
In Kamchatka, for instance, ‘husbands are under the iron rule of their
wives’. Among the primitive Ainu, ‘the wives dictate to their husbands
and make them fetch and carry’. Of the Giak of Sakhalin, a Japanese
traveller writes : ‘In this country it is the custom that the women should
rule over the men; they treat these like servants and make them do all
the work.’ Among the Moi, the most primitive race of Indo-China,
‘nowhere does woman enjoy more consideration and esteem.’

One writer asks how women could possibly have gained such ascen-
dancy over the men. In fact, there is no conceivable process whereby
this could have happened. Not only is the savage strong, but woman is
immeasurably orthodox ; revolt is alien to her nature.

In judging these matters, many writers have been misled by the hard
work done by women to suppose that their status was one of slavery and
oppression. But, in fact, the toil with which they were burdened was
freely and eagerly accepted by them. Generally speaking, it is in those
societies where they toil most that their status is most independent.
Where they are idle, they are as a rule little more than sexual slaves. As
Roscoe says of the Baganda, no woman would remain with a man who
did not give her a garden and a hoe to dig with ; she would return to her
relations to complain of her treatment. Seri women do all the labour of
the community. The toil which woman performs is regarded by her as a
functional division of labour. Indeed, the Australian women are much
attached to the husbands who beat them so brutally, and even the custom
of suttee is nowise resented by women, who often compete among
themselves for the honour of being burned on the funeral pyre. Hero-
dotus tells us that the same competitive eagerness for the honour of
being immolated was the rule among Scythian women, and similar data
can be found among the Maori of New Zealand, the Natchez of North
America and elsewhere.

How then did the revolution take place ? Bachofen, the first to draw
attention to some of the evidence for former female dominance, sug-
gested that the women rebelled in disgust at the promiscuity imposed
by the males. Nothing could be more fantastically impossible. There is
no tendency in women to object to the sexual standards which rule in

THE MOTHERHOOD

her environment, and it is a delusion that she in any way resents poly-
gamy, as many missionaries have noted with surprise. One writes : ‘A
woman would infinitely prefer to be one of a dozen wives of a respect-
able man than to be the sole representative of a man who had not the
force of character to raise himself above the one-woman level.’ And
Miss Kingsley says she knows men, who would rather have had one wife,
driven to polygamy by the women . 1

Similarly women accept promiscuity and sexual freedom where that
is the rule. As the daughters of an African chief said to an explorer:
‘Why do you disdain us ? Are the women of your country prettier . . . ?
We are the daughters of a chief and will not submit to being insulted.’
Similar indignation, when a guest has refused a like offer, is expressed
by Eskimo, Chinook and other women.

The Establishment of Male Domination

Some writers have supposed that the high status of women in certain
communities is the outcome of special economic conditions, such as
their having acquired land or property, and thus that this is a late
development; but the fact that matriarchal organizations exist in
societies which have the barest minimum of economic development,
such as those of the Seri Indians, disposes of this explanation.

Furthermore, there is no evidence of a transition from patriarchal to
matriarchal customs anywhere. On the other hand, in every society,
uncultured or not, where patriarchal usages obtain at the present day,
indications are to be found of a previous higher status of women or of
an actual matriarchal organization . 2

Among matriarchal primitive races, the change to patriarchy is every-
where taking place before our eyes, under the influence of European
contacts. The North American Indians who have adopted paternal
descent, have probably done so within the period of European occupa-
tion. In the islands of Torres Straits the social organization is now
patriarchal, but the guardian of the children at their initiation is not
their father but their maternal uncle. In Nigeria, as among the Malays,
two forms of marriage are found. In one, the children belong to their
father’s clan; in the other, to the mother’s. It is curious that in the island
of Yap, which alone among the Caroline and Marshall groups transmits
property in the male line, matrilocal marriage nevertheless persists, due
to local conditions which do not permit the clans to be sufficiently
segregated.

1 For further examples, sec pp. 210 n.

2 Certain Kwakiutl Indians reckon descent in the male line, but are in the
habit of adopting the crest or totem of their mother’s father. This may be due
to the influence of neighbour clans who reckon descent in the female line.

t.m. — 3*

THE MOTHERS

Sir Edward Tylor compared the two forms of marriage residence
with the custom of mother-in-law avoidance. He found a greater cor-
relation between matrilocy and avoidance than would develop by chance.
Certain tribes, however, display the curious anomaly that, though the
husband takes the wife home to his own tribe, yet mother-in-law
avoidance is carried to a ridiculous extreme, and Tylor concludes that
matrilocal residence must have formerly been customary, adding in
support that, when a native kills game certain parts of the meat are
allotted to the wife’s parents — the duty of supplying game to the wife’s
household is, of course, a well-marked feature of matriarchal law. I
therefore wrote to Mr Howitt, the leading authority on Australian
anthropology, suggesting that inquiry would probably disclose further
evidence of a maternal stage in Australian society. Mr Howitt, after
investigating, replied that this surmise was correct — for instance, in the
event of a war expedition, a daughter’s husband fights on behalf of his
wife’s family and even against his own blood relations. In fact, there can
be no doubt that the patriarchal character of Australian society and the
low status of women are features of comparatively late origin . 1

After Australia, Melanesia is perhaps the area in which the low status
of women is most pronounced. Yet here, too, social organization presents
countless typical matriarchal features, such as the system of maternal
clans, the strict observance of mother-in-law tabus and the special
position of the maternal uncle. Thus, among the natives of the Gazelle
Peninsula of New Britain, though women are dominated by the men
and excluded from religious functions, yet the society is organized
throughout on purely matriarchal principles, descent being matrilinear.
From similar arguments, it would seem that the people of Ticrra del
Fuego were also formerly matriarchal. One of their chief traditions is
that ‘formerly the women exercised supreme domination over the men’,
maintaining this by a system of terrifying apparitions of supposed ghosts,
and the tradition adds that this was terminated by a sort of revolution
in which many of the women were killed. Hence it seems that the change
from matriarchy to patriarchy, which so often does not occur until a late
phase of culture, may also take place at the rudest levels. The low status
of women is such tribes is probably connected with their cultural isola-
tion, for among less-isolated uncultured races the status of women is
generally high.

1 Spencer and Gillen, though they found no instance of matrilocy, also noted
the practice of giving game to the wife’s relations, and concluded that it indicated
a former state in which a man owed allegiance to his wife’s group; in Western
Australia, where the condition of women is particularly degraded, it is neverthe-
less traditional that certain elderly females are invested with the status of
moyram, or grandmother, after which they can no longer be carried off for a
drudge or made the victim of revenge, and they become influential.

THE MOTHERHOOD

On the other hand, those peoples of whom we speak as civilized, such
as the inhabitants of Europe, the so-called Aryan races of India, the
Chinese, the Semitic races of Arabia and their descendants, together
with those who have been converted to their religious systems, are
normally patriarchal, and this association suggests that this is not a
racial peculiarity of certain peoples but the outcome of conditions con-
nected with their state of civilization. It is reasonable to suppose that,
in an earlier stage, these societies too were matriarchal in organization,
or at least less pronouncedly patriarchal than they have been during
historical times, and, when their records are examined, strong evidence
of this is invariably found, as the next chapter will show.

CHAPTER 7

The Matriarchal Phase in
Civilized Societies

The Indian Aryans

Among the Hindus the subordinate position of woman was even more
pronounced than in patriarchal Rome. The laws of Manu assert that
‘in her childhood a girl should be under the will of her father; in her
youth under that of her husband; her husband being dead, under the
will of her sons. A woman should never enjoy her own will. . . .” Her
husband must be worshipped ‘like a god’.

But the oldest Indian records reveal a very different position. ‘Let
a wife’, a Vedic hymn declares, ‘be absolute mistress over her fathers-
in-law, absolute mistress over her mother-in-law; let her be mistress
over her husband’s sisters, let her be mistress over her husband’s
brothers.’ Writers of the epic period were aware of the change. Thus,
Pandy in the Mahabharata says : ‘Women were not formerly immured
in houses and dependent upon husbands and relatives. They used to
go about freely, enjoying themselves as best they pleased. . . . They
did not then adhere to their husbands faithfully ; and yet, O beauteous
one, they were not regarded as sinful, for that was the sanctioned usage
of the times. . . . The present practice of women being confined to one
husband for life hath been established but lately.’

Most significantly, while in the historic and epic periods women were
excluded from religious functions, in the Vedas, on the other hand, they
were present and were even ‘the orderers of the sacrifice’. In the most
solemn function of Vedic religion, the Sacrifice of the Horse, the part they
played was even more essential. 1 At least one of the Vedic hymns was com-
posed by a woman. The erudite Princess Gargi Vakaknavi was renowned
for disputations with the sage Yajnavalkya.

The society depicted in the Vedas is clearly not in a primitive condi-
tion: it appears to be essentially patriarchal, as might be expected of a
pastoral society; it is marked by private ownership of cattle and horses,
by well-defined aristocratic classes of warriors and priests, and a highly-
developed religious cult and literature. It is fantastic to speak, as some

1 See p. 379.

THE MATRIARCHAL PHASE IN CIVILIZED SOCIETIES J]

writers used to do, of the Vedas as ‘the most ancient literature of which
we possess written records’. The Vedas do not, in all probability, go
back much farther than the first millcnium B.C., if indeed as far. The
‘Aryan’ tribes of Hindustan also retained a matriarchal type of social
organization, as is clear from Chinese accounts . 1 Of the people between
the Oxus and the Jaxartes we learn: ‘They hold their women in high
honour, for whatever a woman says her husband invariably agrees to
it” Of the Hu tribe of the Great Get-ti, the historian Hu-Han-Shu
writes : ‘They also take their personal name after their father, but they
take their family name after their mother.’ Thus it would appear that
some of those Aryan tribes had retained a considerable degree of matri-
archal organization. In ancient Indian society itself there are several
indications of the previous existence of matriarchy. Thus, in the Mahab-
harata we are told that a man should avoid marrying a girl of the same
family as his mother, a prohibition still emphasized in Hindu law.The
Pandava dynasty traced its descent from a divine foundress, Pandaia.
In Vedic India the guardian of a woman is her brother. The Rajputs,
the purest representatives of the ancient Aryan conquerors of India,
retain many of the archaic rules which have disappeared among the
Hindus. Thus a woman may only marry into a clan equal to or more
noble than her own, while the first step in any proposal of marriage
must come from the woman or from her family, a characteristic matri-
archal usage. Reverence for women, and especially for a man’s mother,
is a trait of Rajput manners. In Hindu custom, a bride is not at once
permanently transferred to her husband’s family, but at the end of a
month her parents come and take her home, and for five years or until
she has children she lives alternatively in her parents’ or her husband’s
home.

China and Japan

China is the country where patriarchal principles are more strongly
emphasized and women more completely subjugated than anywhere,
and it is generally considered that indications of a former matriarchal

1 Whether or not there was ever, strictly speaking, an Aryan race is a specula-
tive matter. We know, however, that the Medes, the original inhabitants of
Persia, and a neighbouring Scythic tribe in what is now Turkestan, called them-
selves thus. The Medes called themselves Ariyas, and the eastern portion of
Persia was called Ariana — Iran is probably a variant of it. The peoples included
under the denomination Scyths were an Iranian race, and spoke Iranian languages
and had similar religious conceptions. In India the term ‘Arya’ was used with
reference to the three upper castes rather than as an ethnic designation. With
one important exception, no Hindus called themselves Aryas. The tribes which
the Romans called the Great Gctae inhabited the deserts of Scythia, and seem
to have been the same as the tribes in India called Jats.

THE MOTHERS

state are completely lacking. The early social history of China is obscure,
but nevertheless there are indications that China is actually no excep-
tion. Thus, while emphatically patriarchal, the Chinese are exogamic.
A man is strictly forbidden to marry a woman bearing the same clan-
name as himself. The names of these clans are made up of the signs
meaning ‘woman’ and ‘birth’, and their combination thus means ‘born
of the same woman’ or ‘one woman’s brood’. Chinese marriage customs
are essentially identical with those of the Mongol and nomadic Tartar
populations of Central Asia, except for the more emphatic consolidation
of marital power. The continuity of evolution between the ruder tribes
and the Chinese can be clearly traced. Thus, among the Mongols and
Tartars, as in China, the bride is purchased. Nevertheless, the bride is
obliged to supply the house itself, which, as Robertson Smith pointed
out, is a survival of the time when it was in the wife’s house that the
husband took up his abode. The relationship between this rule and the
matriarchal order of society has been noted by the Chinese. As one
writer says : ‘The girl’s dwelling and trousseau all come from her own
family; hence the custom of counting genealogies from the female
side.’

It follows that such usages must have been preceded by a stage in
which marriage was matrilocal, a conclusion supported by the fact that,
at the present day, it is only after living several days in the house of his
parents-in-law that the bridegroom takes his bride to his own home.
Throughout China it was once customary with the aristocratic classes
that the wife, after a short residence in her husband’s home, should
return to her own house for several months. Moreover, complete matri-
local marriage is still sometimes practised at the present day, and
Chinese law provides for the case of a husband who takes up residence
in the home of his wife. Such a husband generally assumes the wife’s
family name instead of her taking his. It appears from the language of
Li Ki that such marriages are a very ancient practice which has only
recently become rare. In many districts there are anti-marrying leagues
among the girls, who refuse to accept the complete effacement which is
the lot of a Chinese wife. In Lung Kong they refuse to marry except
on the condition that the husband go to the wife’s home to live. In
Yunnan the women appear to retain a degree of independence, and as
a preliminary to marriage the bridegroom comes to the girl’s house and
knocks. His intended then asks who is there and, on hearing his name,
asks if he wishes to come to her house and stop with her. The children
take the wife’s family name.

Such facts cannot be interpreted as indicating an evolution from
patrilocal towards matrilocal usages. According to a Chinese tradition,
the present form of marriage was instituted by Fu Hi, the legendary

THE MATRIARCHAL PHASE IN CIVILIZED SOCIETIES 79

founder of Chinese civilization, previous to which children did not know
their fathers and sexual relations were promiscuous, which is the usual
expression used by patriarchal peoples when referring to matriarchal
marriage. Fu Hi himself had no father, and the same is true of all the
legendary and of many later Chinese emperors. Early Chinese history
offers many examples of powerful and masterful empresses who either
ruled as regent or forcibly seized the reins of government. Famous
among them are Lu-Kao-Heu (187 bc), Sing-Zche (ad 1) and Hou-
Zche (ad 15). The position of the women of the royal family, where
archaic usages survived, constitute an exception to the rule concerning
the position of women in China. When a minor was emperor, the
dowager empress ruled for him. She selected the empress — his chief
wife — his eight queens, and also his ministers, generally appointing her
own brothers. Though, in later times, only the Son of Heaven could
perform the sacrifice to Sanf-Tien, and scandal was caused when the
adventuress, Hou, arrogated to herself this privilege, yet one of the
earliest references to that sacrifice reads: ‘The Empress Kiang-Yun,
together with the Emperor, offered a sacrifice to Shang-Ti.’ It is clear,
therefore, that the status of women in religious matters must have
undergone a transformation. As late as the third century of our era,
women could hold office and exercise administrative functions, a right
which did not completely disappear until the eighth century.

Turning to Japan, it might seem, at first sight, that the position was
identical with that in China. Until recently the position of the Japanese
upper-class wife was quite as servile as that of the Chinese wife. Yet,
it is clear even from existing customs that such usages were mere
imitations of the Chinese conception. For instance, instead of marriage
being regarded as a solemn contract which even death cannot dissolve,
there is complete liberty of divorce for both parties : 33 per cent of
Japanese marriages are said to end in divorce, and both parties usually
marry again without delay. When a daughter is the sole heiress, she
continues to live in her parent’s house, instead of going to her husband’s,
and her husband assumes her name and becomes an adopted son of her
parents.

The features of marital relations which resemble those of China are
much more pronounced among the upper-class than among the com-
mon people. Among the latter the wife is treated with deference, and
exercises her right to dismiss the husband on the slenderest grounds.
This, together with the matrilocal custom, suggests that the Chinese
usages are recent adoptions. Patrilocal marriage did not, in fact, come
into use in Japan until the fourteenth century. Previously, the husband
merely paid occasional visits to his wife, who remained in her parental
home. The Japanese family was entirely uterine, the children taking

THE MOTHERS

the name of the mother. A ceremonial survival still exists in the obliga-
tion of a newly-married couple to spend a night or two in the bride’s
home. Still older records represent an even more pronounced degree
of the same condition of things. There was, properly speaking, no
marriage — that is to say, no contract. Sexual cohabitation constituted
marriage, and in older Japanese literature it is not possible to distin-
guish the relations of a wife from that of a mistress or concubine.
Powerful chiefs are described as having a wife in every island and on
every headland.

The Semites

The patriarchal clans of the Hebrews have always stood for the very
type of patriarchal organization, yet it was from a study of their customs
that Robertson- Smith illustrated the transition from matriarchal to
patriarchal institutions. Semitic kinship terms refer, as among most
peoples, to maternal rather than to paternal descent, and the clan or
tribe to which a man belongs is frequently spoken of as his ‘mother’.
The Arab genealogists of the time of the rise of Islam, when patriarchy
was strongly established, were at pains to exhibit Semitic descent as
patriarchal, but they were often compelled to recognize that the older
clans were metronymic. Among the Hebrews who had left the penin-
sula many centuries before the Arabs, the same situation had arisen
long before, and they endeavoured also to interpret their genealogies
in terms of a male descent. Yet the Jewish rabbis themselves, at a com-
paratively late date, acknowledged that the ‘four Matriarchs’, Sarah,
Rebecca, Rachel and Leah, had occupied a more important position
than the ‘three Patriarchs’, Abraham, Isaac and Jacob. According to
Robertson- Smith, the tribe of Levi was originally metronymous, being
the tribe of Leah for whom the husband Levi had to be invented.
Israel itself, the tribe which gave its name to the whole nation, was
originally the tribe of Sarah, Israel being her son. Even in patriarchal
times, women built cities, i.e. founded families. Sherah ‘built Beth-
Horon the nether, and the upper, and Uzzen-Sherah.’ (i Chronicles
vii. 24) 1 . With the early Hebrews the regular practice was for the man to
‘leave his father and mother and cleave to his wife’; that is, to take up
his abode with his wife’s clan. Isaac takes it for granted that Jacob,
when he marries, will dwell with his wife’s people. In fact, Jacob lives
twenty years in the home of his wives and, when he departs by stealth,
Laban pursues him and tells him that he has no right to take them, or
even his own children, away, and claims them as belonging to their
mother’s father. Samson’s wife remains with her own people. Joseph’s

1 The Jewish practice of marrying a father’s daughter or his sister (Genesis xi.
29; Exodus vi. 20; Ezekiel xxiii. 11) also points to former matrilocal usages.

THE MATRIARCHAL PHASE IN CIVILIZED SOCIETIES 8l

children by his Egyptian wife have to be adopted before they can be
regarded as belonging to his tribe.

Matrilocal marriage passed away when the Jews settled in Canaan*
but survives to this day among the Bedawi of Arabia. To the true
Bedawi, marriage is almost always matrilocal. ‘The wild men’* says
Burton, ‘do not refuse their daughters to a stranger, but the son-in-
law would be forced to settle amongst them . 5 Even after the introduction
of patrilocal marriage, contracts were sometimes drawn up stipulating
that the wife should remain with her own people.

Ammianus Marcellinus noted that in his day the common marriage
ceremony among the Arabs consisted in the woman's presenting to the
man a spear and a tent. The tent has always been regarded by the Arabs
as the property of the woman and, indeed, as a synonym for the woman.
An Arab author mentions that the women of Jahiliya could dismiss
their husbands by merely turning their tent round, a custom which still
persists. Among the Bedawi of Mesopotamia a special tent called ‘hofah 5
is erected for the newly married couple. The ordinary expression for
‘to get married 5 in Jewry was to ‘go into the “huppah 55 5 , i.e. the tent.
The marriage was not concluded until this had been done. When the
practice of the husband visiting his wife in her tent gave place to patri-
local marriage, he did not bring his bride immediately into his own
house, but a special tent, presented by the bride’s family, was erected
to represent the woman’s home. ‘Huppah 5 is still the ordinary term for
the marriage ceremony among the Jews and, until quite lately, was
represented in every Jewish marriage by a canopy under which the bride
proceeded to the synagogue.

As we should expect from the former existence of matrilocy, there
are indications that formerly the position of women, both among the
Jews and in Arabia, was high. The oldest existing fragment of Hebrew
literature represents the Hebrew tribes as led by a woman (Judges
iv-v). Among the Arabs likewise many of the judges were chieftainesses.
Queens, from the Queen of Sheba to Zenobia, occupy a prominent
place in Arab history. In every one of the references to Arabian king-
doms in the historical inscriptions of Assyria, it is with a queen that we
have to do. The position of those queens appears to have been exalted.
Their husbands were merely their consorts. ‘The female sex rules
among the Sabians , 5 says Claudian, ‘and a large proportion of the bar-
barians is under the armed domination of queens . 5 Women in ancient
Arabia were commonly the owners of wealth, possessing large flocks
and herds; their husband commonly acted as herdsman. Mohammed
himself was able to carry out his mission thanks only to the wealth which
he acquired from his first wife. In early and pre-Islamic literature we
come upon women like Mawia bint Afzar, who changed her husband

THE MOTHERS

whenever she pleased ; but, though courted by the most famous warriors
and poets of the age, her choice was never frivolously determined. There
is nowhere to be found a nobler type of dignified yet unfettered, free yet
self-respecting, womanhood than the women of ancient Arabia up to
the time of ancient Saracenic civilization. The women were always close
behind the warriors in battle, and more than once rallied them to
victory. Many an Arab woman personally led the men to battle. As late
as the time of Rashid, Arab maidens fought on horseback and com-
manded troops, and royal princesses clad in mail fought under Mansur
against the Byzantines. Yet those proud Amazons were not barbaric
viragoes, but cultivated beauty, grace and elegance and all the accom-
plishments of their age. The record of their sentiments is contained in
their own impassioned poems, which form a considerable proportion of
pre-Islamic poetry. The tent of many an Arab woman was the scene of
tournaments of song. Like archaic Greece, ancient Arabia had its seven
sages, but they were women.

Ancient Egypt

The conservative society of Egypt never lost its essential matriarchal
character, though continuous progress towards patriarchal institutions
is clearly traceable through the various phases of Egypt’s four-thousand-
year career. We shall have to consider elsewhere the rules of succession
to the throne of the Pharaohs, and we shall see that the functions of
royalty were transmitted in the female line. While every Egyptian prin-
cess was born a queen and bore the titles of the office from the day of
her birth, a man only acquired them at his coronation, and could do so
only by becoming the consort of a royal princess. It was in the queen
that mystic or divine virtue resided.

Such features are substantially identical with those obtaining in all
other African kingdoms. Royal families naturally tend to preserve a
more archaic constitution than the families of ordinary people, and
these matriarchal features would not in themselves be sufficient ground
for inferring the matriarchal character of Egyptian society as a whole.
Nevertheless, scanty as our information is we have enough to see that
Egyptian social life generally was in the highest degree matriarchal.
Dr H. R. Hall says that the constitution of Egyptian society was
characterized by ‘a distinct preservation of matriarchy, the prominent
position of women and a comparative promiscuity in sexual relations’ .
Professor Mitteis confirms that Egypt was immemorially a land of
matrilinear descent, a usage which continued in Christian times until
the seventh century. ‘The maternal uncle is often named as important.
The father of the mother was more important than the man’s own father.’
In consequence, there were no illegitimate children in Egypt, a child

THE MATRIARCHAL PHASE IN CIVILIZED SOCIETIES 83

born out of wedlock enjoying the same rights as one born in marriage.
All children belonged to the mother. The Nomes, or primitive totemic
clans, of which the nation was formed, were maternal clans or mother-
hoods, and their headship was transmitted through women.

Flinders Petrie observes : ‘The family in Egypt was based on a matri-
archal system. . . . The house and property went with the women and
daughters.’ Sir Gaston Maspero says: ‘The Egyptian woman of the
lower and middle class was more respected, more independent, than
any other woman in the world.’ The husband had no power as head of
the household, and the word ‘husband’ is only found in marriage con-
tracts after the reign of Philopator. Marriage was matrilocal, and where
there were two wives each remained in her own house, the husband
visiting them in turn. ‘As late as the XIXth Dynasty, there was still
surviving the idea that a man was only a boarder in a woman’s home.’
All landed and house property was in the hands of the women, and if a
man built a house it passed immediately to his wife.

Marriage does not appear to have been associated with any religious
ceremony, but was essentially an economic transaction and was made
the subject of a written contract. We possess hundreds of these contracts
dating from the Ptolemaic period. In them the woman imposes her
conditions on the man. Thus : ‘If I leave thee as husband because I have
come to hate thee, or because I love another man, I shall give thee two
and a half measures of silver, and return to thee . . . the bride-gift.’ In
a love poem of the period of Rameses II, addressed by the lady to her
beloved, she opens her heart thus : ‘O my beautiful friend ! My desire is
to become as thy wife, the mistress of all thy possessions !’ The eagerness
which Egyptian husbands seem to have shown in making over their
property to their wives was probably due in part to the fact that, accord-
ing to matriarchal usage, it would otherwise have passed not to their
own but to their sisters’ children. Thus, by a curious paradox, the anxiety
to secure patriarchal succession contributed to accentuate the economic
power of women. To the same desire to combine inheritance in the male
line with the matriarchal organization of the family was also doubtless
due the practice of marrying sisters, which appears to have been more
prevalent in Egypt than among any other people of which we know. The
majority of marriages were between brothers and sisters as late as the
second century ad in some districts.

Generations of Egyptologists have treated with contempt the state-
ment of Diodorus Siculus that under the marriage agreement the hus-
band appertains to the wife and must obey her, but this is now known to
be strictly accurate, and obedience is urged on the husband as a duty in
‘the oldest book in the world’, the ‘Maxims of Ptah-Hotep’, dating from
about 3200 bc. Indeed, the marriage contracts which we possess go even

THE MOTHERS

further than does Diodorus ; the tone of servility which pervades them is
almost incredible. 'I acknowledge thy rights as wife/ says one of them,
‘from this day forward I shall never by any word oppose thy claims. I
shall acknowledge thee before anyone as my wife, but I have no power
to say to thee: “Thou art my wife.” It is I who am the man who is thy
husband. From the day that I become thy husband I cannot oppose
thee, in whatsoever place thou maycst please to go. I cede the . . . [here
follows a list of possessions], that are in thy dwelling. I have no power to
interfere in any transaction made by thee, from this day. Every docu-
ment made in my favour by any person is now placed among thy deeds,
and is also at the disposal of thy father or of any relatives acting for thee.
Thou shalt hold me bound to honour any such deed.’

Can we wonder at the gibes of the Greeks at the ‘topsy-turvy world’
of Egypt and its hen-pecked husbands ? It is not surprising therefore
that the position of women in society was equally prominent. Whether
married or single, from the earliest age a girl had the fullest legal rights
and could enter legally into any transaction. Guardians were entirely
unknown in Egypt. ‘No people, ancient or modern/ says Max Miller,
‘has given women so high a legal status as did the inhabitants of the
Nile valley.’ Herodotus declares that, ‘the women go in the market-
place, transact affairs and occupy themselves with business, while the
husbands stay at home and weave’. Even the role of priestess was open
to a woman. We have full details of the career of a woman who, begin-
ning as a clerk in her father’s office, was promoted to various adminis-
trative posts, and became governess of the Fayum, and, what appears
more remarkable, commandcr-in-chief of the western military district.
To this was added later the governorship and the military command of
Kynopolis and the eastern frontier, and she became one of the most
important and wealthiest personages in the realm.

The continuous restriction of these privileges can be traced. During
the first dynasties the number of women exercising the function of
priestesses arc numerous, but there is not a single example after the
Xllth Dynasty. Thus it would appear that, as in many primitive tribes,
the first step in the limitation of the status of women is to take over from
them the monopoly of religious and magic functions. By Ptolemaic times,
it is the husband not the wife who claims complete freedom of divorce,
and husbands are found administering the estate of the wives.

The JEgean and Greece

The marvellous remains of Krete provide eloquent testimony to the
matriarchal character of Kretan society. The enormous predominance
of female over male figures is without parallel. Kretan divinities are
almost exclusively feminine, as are the innumerable votive figures.

THE MATRIARCHAL PHASE IN CIVILIZED SOCIETIES 85

Women alone figure as priestesses in religious ceremonies. These
characteristics arc equally marked from ancient Neolithic times to the
height of Minoan culture. ‘It may be admitted without reserve’,
remarks Dr Mosso, ‘that at this epoch, that is to say, about 1500 bc,
Minoan religion had preserved its matriarchal character. The supre-
macy of women in religion was thus maintained until Mykenean times.*
Secular art also speaks for the predominance of women. There are
numerous representations of men in Minoan art, but they are all en-
gaged in subordinate occupations : cup-bearers, pages, musicians,
harvesters, soldiers and sailors. Not once does it depict a king, prince or
hero, or show a man in a position of domination. The female figures, on
the other hand, invariably exhibit an attitude of self-possessed indepen-
dence. Minoan women are seen mixing freely with men at festivals,
riding in chariots driven by female charioteers. ‘It is certain’, remarks
Dr Hall, ‘that they must have lived on a footing of greater equality with
the men than in any other ancient civilization.’ That the state was
governed by women is suggested by Klidemos, who tells us that after
the death of Deukalion the throne passed to Ariadne, who concluded a
treaty of peace with Athens.

Marriage in Krete was matrilocal, as appears from Strabo, and from
the laws inscribed on the walls of the temple of Gortyna, which contain
many traces of older matriarchal usages. Thus the mother’s brother
occupies an important position and is responsible for the bringing up of
the children. A woman, on marriage, retained full control of her pro-
perty and, like the husband, had the right of divorce at pleasure. A man’s
property passed to his children, but in the absence of children his sister’s
children might be heirs.

Krete was the most brilliant focus of a culture which, at that time,
was common to all the peoples of the Aegean. The various peoples of
the Aegean were culturally, linguistically and doubtless ethnically one.
Lykians, Karians and Lydians are all one race, and what we know of
them confirms their matriarchal character. Thus Nicholas of Damascus
tells us that ‘amongst the Lykians the women are more honoured than
the men’, and that not only kinship but property was transmitted in the
maternal line. Of the Lydians we are told : ‘The men are subject to
female domination. The throne went with the queen and not with the
king.’

Probably the same race that peopled the Aegean also peopled the
greater part of the Mediterranean coast. Every advance in our know-
ledge tends to confirm the view that the Neolithic race of the Mediter-
ranean came from Africa at a time when land-bridges still spanned the
inland sea. The Berbers and Tuareg would thus be the surviving
African relatives of the race which gave birth to European culture, and

THE MOTHERS

the similarity of the scripts of Minoan Krete with the archaic script
preserved by Tuareg women seems to confirm this.

Female figures are equally predominant in the Neolithic art of the
whole Mediterranean region ; there is not a single male idol to be found
before the Bronze Age. Dr Rodenwalt says that the prominence of
women is even more pronounced in continental Greece than in Krete,
and points out that, whereas in Kretan frescoes women are assisted by
men, in the Mykenan palaces of the Peloponnesus ‘we see, everywhere,
women and women only taking part in religious cult’. At the opposite
end of the Mediterranean the evidence of matriarchy is even more
emphatic. ‘Among the Cantaberians’, says Strabo, ‘the men bring
dowries to the women. With them the daughters alone inherit property.
Brothers are given away in marriage by their sisters. In all their usages,
their social condition is one of gynaecocracy.’ Until quite recently the
custom lingered in the Pyrenees that all property passed to the eldest
child, son or daughter, and, if the latter, the husband took up his resi-
dence at the house of his wife, but acquired no rights over her property.

It has frequently been suggested that, while the aboriginal or Pelas-
gian populations of the Aegean were matriarchal in their social organiza-
tion, the invaders, or Hellenes, brought with them ‘as a precious posses-
sion, the patriarchal system’. It is extremely doubtful, however, whether
there exists any fundamental ethnical difference between the autoch-
thonous inhabitants and the invaders. The Dorians, whom alone
Herodotus calls ‘Hellenes’, came from the north, but there is no evi-
dence that their habitat was any farther than the southern shores of the
Danube, a region where we find the Neolithic culture of Butmir iden-
tical with that of Krete . 1

Be that as it may, the Spartans appear to many ‘the embodiment of
the specially Hellenic elements in Hellenism’. Their social organization
was purely matriarchal. Their customs were so similar to those of Krete
that it was believed at the time that they had been borrowed from the
Kretans. Spartan women were entirely unrestricted in their social and
sexual relations. Virginity was not demanded of a bride. Children born
out of wedlock were called ‘virgin born’, and were regarded as equal to
those bom in wedlock. At the time of Argesilaos, the number of them
exceeded those born in regular wedlock. The Spartans practised fraternal
polyandry, and their marriage was matrilocal. In the Spartan version of
the story of Penelope, when she follows Odysseus home, she is repre-
sented as breaking the custom. Plutarch says that the Spartan women

1 The assumption that the Greeks, or ‘Achseans’, were a separate race from
the Aegeans is based upon the Aryan theory, a speculation which has now
crumbled to pieces. It is not without significance that Dorian invasion was
invariably spoken of by the Greeks as ‘the return of the Heraklides’.

THE MATRIARCHAL PHASE IN CIVILIZED SOCIETIES 87

were ‘the only women in Greece who ruled over their men’. They were
commonly consulted on political questions, and could inherit and bestow
property, nearly all the property in Sparta being in their hands.

The Lokrian colony of Cape Zcphiros was also regarded as having
preserved more archaic usages than other Hellenic states. Of them we
are told : ‘All fame and honour attaching to descent is derived through
the women, not through the men.’ Ritual sexual licence, quite opposed
to the sentiments of most Greek communities in historical times, sur-
vived among them until a late date. The view that the Greeks were
originally matriarchal is confirmed by traditions of the historical period,
when it was held that the status of women in the form of marriage had,
in primitive times, been entirely different. It was said that in primitive
Athens ‘at one time, because of the general promiscuity, men did not
know their own fathers’. Marriage was said to have been instituted by
Kekrops, who was therefore named ‘diphues’ — ‘of a double nature’ —
for before him children had a mother but no father. Aeschylus in his
Eumenides represents the change from the ancient law to that of the new
gods as chiefly manifested in the way of viewing maternal and paternal
kinship. The tradition of the contest between the old and the new gods
to which the Eumenides referred, gives an account of the changes in the
status of the women. The famous contest between Athene and Poseidon
for the possession of the city was decided by the votes of the Athenian
citizens. The women voted for Athene, the men for Poseidon. When the
victory went to Athene, Poseidon vented his anger by flooding the land,
and in order to pacify him, it was ruled that the women should be dis-
franchised, that no child should receive the name of its mother and that
women should no longer be regarded as Athenian citizens.

While such traditions are of course mythical, no mythologist nowa-
days doubts that such myths indicate an actual conflict between native
and foreign cults. It would be difficult to imagine why a strictly patri-
archal people should devise a theory of their primitive matriarchy. A
considerable body of evidence confirms these traditions. As we shall see
elsewhere, in primitive Greece the religious functions connected with
agriculture were exercised chiefly if not exclusively by the women. The
chief civic function in this connection continued in historical Athens to
be exercised, not by the male magistrate or ‘archon’, but by a special
female functionary, the ‘queen archon’, assisted by a council of matrons.
Gods and heroes are commonly referred to in Greek genealogies by the
names of their mothers, as ‘Apollo, the son of Leto’, ‘Dionysos, the son
of Semele’, and so on. Such gods and heroes were in fact, ‘virgin-bom’;
that is, they were in the same case as the Athenians are reported to have
been before Kekrops instituted marriage. Jason is expressly asserted to
be ‘virgin-born’. The Argonauts, or Minyans as they are commonly

THE MOTHERS

called, all trace their descent through women to a common ancestress,
Minya, or to Klymenes, mother of Jason. Early Greek genealogies are,
in fact, little else than catalogues of women. The relationship between
Theseus and Herakles, to which so much importance was attached in
archaic tradition, was traced through women. In primitive Greece the
women gave their names, not only to their children but to their clans
and tribes. Thus, the Athenians claimed to be descended from Atthis,
the daughter of Kranaos ; the Spartans from Sparta, the daughter of
Eurotas, and so forth. The grammatical ending of Greek family and
tribal names in ‘-ida’, is a purely feminine form. Later traditions sub-
stituted obscure male eponyms for the original eponymae. Thus the
Ionians were supposed to be called after a certain Ion, but their true
eponym is revealed by the fact that the Ionian sea was understood to
be named after Io. Similarly the Dorians were more probably derived
from the lunar goddess Doris than from an obscure Doras, a son of
Helen. Both Ionians and Dorians traced their descent from Helen, the
daughter of the Moon, and there can be, I think, little doubt that she
was the true ancestress of the Hellenes.

Greek thinkers in the classical age held that the mother had little
or no share in the process of procreation, that the seed proceeded from
the father, and the mother’s womb was but a suitable receptacle to
protect it. The mother, it was said, was little more than a nurse; the
father was, strictly speaking, the sole progenitor. This view contrasts
not only with older usages but with the very structure of Greek speech
and legal customs. In Homer a sharp distinction was drawn between
a uterine brother, ofioyaorQioq, and a brother on the father’s side,
onaxQoc , , and the former relationship is invariably emphasized as more
important. Moreover, in contradiction to this ingenious patriarchal
theory of procreation, the ordinary appellation for brother continued
to be ‘adelphos’, which means ‘from the womb’, and is therefore a relic
of the time when the relationship of a brother was reckoned on the
mother’s side only. The distinction persisted in the most concrete
manner in historical times, for, according to Athenian law, a man was at
liberty to marry his half-sister on his father’s side, but was forbidden
to form an incestuous union with his half-sister on the mother’s side.

In princely houses, of which alone traditional records have reached us,
it is the women who transmit both titles and property. They remain in
the maternal homes, and the sons regularly depart to marry in some other
town princesses whose titles they share. This is the form of marriage
which Alkinoos proposes to Odysseus. T should wish’, he says, ‘that
so goodly a man as thou . . . would take my daughter to wife and be
called my son and abide with me.’ Gilbert Murray thus briefly sums the
social conditions in the heroic age of Greece : ‘House property belonged

THE MATRIARCHAL PHASE IN CIVILIZED SOCIETIES 89

to the woman, and descended from mother to daughter. The father
did not count — at least not primarily — in the reckoning of relationship.
He did count for something, since exogamy, not endogamy, was the
rule. The sons went off to foreign villages to serve and marry women in
possession of the land there. Their sisters, we have reason to believe,
generally provided them with dowries.’

In Athens, in historical times, it was the custom that after an Athen-
ian husband had removed his wife to his own home and spent the wed-
ding night there, the couple should return on the following night to
the bride’s home and sleep there — a custom observed by many peoples
who formerly practised matrilocal marriage; for instance, the Baila of
Rhodesia. In fact, Athenian marriage never became thoroughly patrilocal.
The Athenian wife, though she removed to her husband’s home, never
became legally regarded as a member of his family. Her father could, at
any time, take her away and bring her home or marry her to another
man. If she had no father, her next-of-kin or legal guardian could do
the same. A wife had no claim on any of her husband’s property except
her dowry, and when he died the widow returned to her own people.
It is, I suggest, impossible to conceive such laws as having developed
in a state of society where the tradition was for the husband to transfer
the wife to his household, and to become the founder of a patriarchal
society.

In the islands of Greece, little touched by the passage of events, the
traveller often comes upon scenes that answer in every detail descrip-
tions in the Iliad and the Odyssey. In the island of Kythnos, at the
present day, it is the invariable custom for a husband to take up his
abode after marriage with the family of his wife.

The adaptation of the laws of matriarchal society to patriarchal ideas
is no less clearly exhibited by the legislation concerning inheritance;
Athenian law in this respect being very similar to the Kretan law. An
heiress must marry her nearest male kinsman, except in the forbidden
degrees, starting with her father’s brother, if any. This is clearly a pro-
vision to ensure that property shall remain with the male members of
the family without going so far as to make it transmissible in the male
line only. Athenian law was very similar but even stricter, for if the
woman was already married at the time she inherited, the male heir
had a right to take her away from her husband, whose marriage became
null, and to marry her. Professor Ridgeway feels that this points ‘to a
time when all property descended through women’.

No contrast could be more glaring than that presented by the position
of women in the Homeric poems and that found in historical Greece,
which, as will be noted later, was beyond all comparison degraded.
Throughout the Odyssey it is the women who direct, counsel and protect

THE MOTHERS

the men. Wherever Odysseus goes, he comes upon a queen or queen-
goddess ruling the land alone, or with a subordinate consort. The royal
office is transmitted by dynastic incest. He recites catalogues of women,
giving the female genealogies of the Minyan and Aiolian houses. So
marked is this that Richard Bentley, the scholar, declared that the
Odyssey had been specially composed for women, and Samuel Butler
had the whimsical idea that it must have been written by a woman . 1
The Odyssey recalls a society in which women held a great place . 2
Klytemnestra, Alcestis, Kassandra, Medea, Polyxena, Hermione,
Antigone differ completely from the sequestered Greek wife, artificially
stunted in mind, who was not even permitted to witness the represen-
tations of her ancestresses on the stage.

The Teutons

Passing to the barbaric nations of Europe, everyone is familiar with the
account given by Tacitus of the influence wielded by women among the
Teutons. In the oldest Nordic and Germanic documents persons are
often referred to by the names of their mother without mention of the
father. The Lombard nation traced its descent from a woman, Gambara,
the father not being mentioned. Tacitus notes that ‘the sons of a sister
have the same position as regards their uncle as with their father. Some
even consider the former as the stronger tie\ Down to a late date the
uterine relationship was regarded as more important than the paternal.
In the time of Tacitus the organization of the Teutons was patriarchal,
for a man transmitted his property to his own children. This was com-
bined with purely matriarchal juridic usages. Thus, according to the
laws of the Thuringians, if a man died without children his property

1 Note that the lines in which Alkinoos and Arete are unambiguously stated
to be brother and sister are immediately followed by a genealogy which repre-
sents them as uncle and niece — a concrete illustration of the adaptation of
matriarchal to patriarchal ideas.

2 It must be concluded that the Iliad and the Odyssey , in their present general
form, were composed not earlier than the eighth century b.c., and probably in
the seventh by the remodelling of much older material. Thus, they cannot be
regarded as affording a contemporary picture of the world they depict. In many
respects, they represented the world they depict about as accurately as Shake-
speare’s Hamlet represents fourth-century Denmark. Thus, among Homeric
heroes iron is not only in general use, but its use is so familiar that it has passed
into proverbial expressions ; whereas Mykenean Greece belonged to the Bronze
Age and iron was completely absent, except in the form of very rare finger-
rings. Again, Homeric heroes are cremated. Mykenean Greece invariably buried
its dead. And so on. Homer seems deliberately to have archaized his material;
he carefully avoids, save in one ambiguous instance, mentioning writing,
although writing had been in use in the Aegean from time immemorial. At the
same time, he adapts myths such as that of Penelope to suit the moral notions
of his time.

THE MATRIARCHAL PHASE IN CIVILIZED SOCIETIES 91

passed to his sister, or failing her, his mother. Similarly, in Burgundy
a man’s property, in the absence of children, went to his sister and, fail-
ing her, to his brother. Of the famous Salic laws, there are ten somewhat
different redactions. In the four oldest a man’s heirs, after his son, are
his mother, his brother, his sister and his mother’s sisters. In none is the
father mentioned; in one only is the father’s sister named, and she owes
precedence to the mother’s sister. The later redactions show Roman
influence, and a gloss informs us that ‘the laws which were observed in
pagan times are no longer valid, for according to them many persons
were deprived of their rights’. It was not until the end of the sixthTcen-
tury that the legal equality of the father’s and of the mother’s relations
became fully established. Ample evidence confirms that property and
titles, at least in royal houses, were transmitted through the women.
Saxon aspirants to the throne did not consider that they had fully estab-
lished their claim until they had married the queen.

Thus Hermigisil, King of the Varini, on his deathbed, begs his son
Radger to be sure and marry his stepmother in accordance with their
ancestral custom. Edbald, King of Kent, does likewise; as does Ethel-
bald, King of the West Saxons, who marries the widow of his father
Ethelwulf. The West Saxon Queen Seaxburg, however, preferred to
retain the kingdom for herself. Similarly, Canute the Dane, after having
overthrown Ethelred, sends for the queen, an old woman who was
then living in Normandy, and does not regard his usurpation as com-
plete until he has married her. Hamlet’s uncle, Feng, obtains the
Danish crown in the same manner; and Wiglet killed Hamlet in order
to obtain possession of Hamlet’s wife and of the kingdom. Similarly
among the Scandinavians the kingdom passed to the daughters and to
their husbands as late as the eighth century; and it is usual in Nordic
records for the kingdom to be inherited through a man’s mother, or by
marrying the queen or a royal princess.

We have no express information as to whether marriage was patri-
local or matrilocal among commoners, but we know from Procopius
that the Gothic tribes were familiar with matrilocal marriage.

The Celts

Though we have no systematic account of the Celts as a whole, we know
a great deal about the Irish Celts, and there appears to have been great
uniformity in the usages of all Celtic peoples, whether they dwelt in
Ireland or on the banks of the Po or the Tagus. Irish traditional narra-
tives show that marriage was essentially matrilocal. Even Cuchulainn,
mightiest of all the heroes, was constrained to go and live with the fairy
queen, Fand, who had wooed him. Throughout, the mistress or wife
retains superiority. ‘She chooses whom she will and is no man’s slave.

THE MOTHERS

Herself she offers freely but she abandons not her liberty/ Irish and
Welsh heroic groups are named after the mother and, especially in the
older strata of tradition, heroes are commonly metronymous. Ireland
and Scotland derive their names froiri eponymae, Erin and Scota, and
the earliest settlers in Ireland are represented as having been women.
The queens and princesses of Irish traditional history are pictured as
viragoes, of whom their husbands generally stood in dread.

Among the Piets, property was transmitted exclusively matrilineally, a
man’s estate passing to his sister’s children. ‘It is in the right of mothers
they succeed to sovereignty and all other successions’. Accounts of early
Scottish history confirm what we should expect from this. Women did
not leave home on marriage, a usage of which traces survive in the High-
lands even today. Tt is a practice among the better sort in these days
for the bride to remain with her parents for some weeks.’ The accounts
of the Book of Leinster and other documents state that matriarchal
organization was imposed upon the Piets by the Irish Gaels after a
victory. This, of course, is unlikely, but implies that the Irish themselves
had the same social customs, for they would hardly impose upon other
peoples usages foreign to themselves. Throughout the literature of the
Irish and British Celts heroes are represented, as in the archaic Greek
sagas, as leaving their homes to seek in some foreign tribe an heiress
whom they will marry and with whom they will share dominion over
her estates. The ancient laws of Ireland and of Wales lay down that
when a daughter is married to a stranger her son shall inherit the estate,
and the principal personage after the chieftain is invariably his sister’s
son, not his own. There can be no doubt that matriarchal succession
was the immemorial rule with all Celtic-speaking peoples.

Ancient graves of the Bronze Age in Britain show that women were
buried with at least as much pomp as the men, and Tacitus reports
with surprise that Britons, brought before the Emperor Claudius to
make their obeisance, went straight to the Empress Agrippina and did
homage to her.

The little that we know of pre-Roman Gaul is also consistent with
the matriarchal interpretation. Several accounts testify to the influence
of women, who took part in a tribal council. Almost the only account
of their marriage customs we have is one given by a Greek merchant
who witnessed a wedding. He was surprised to be invited to a banquet
to which a large number of marriageable young men had been invited.
After it the bride entered and surveyed the assembly, then signified
her choice by presenting a cup to the man she had selected as husband.
Similar matriarchal usages are found in various parts of the world;
matrilocal marriage is found nowadays in some secluded mountain
districts of Savoy and is known as ‘goat marriage’, an allusion to the

THE MATRIARCHAL PHASE IN CIVILIZED SOCIETIES 93

practice of bringing the he-goat to the female* while cows and ewes are
taken to the male.

Rome

Rome stands for the very type of the patriarchal organization of society*
and* in large measure* it is from her that our own patriarchal organiza-
tion has been derived. Modern jurists and historians have analysed
the Roman tradition as it is embodied in the interpretations of Roman
jurists* and have derived from it the patriarchal theory of society, which
has been worked out with great learning, notably by Sir Henry Maine.
It assumes that the patriarchal family as we know it is not only the
unit of human society but the primitive and original form of the human
social group* and that all other groups* such as the clan or gens* the tribe
and, finally, the city and state* have been formed by the aggregation of
patriarchal families and ‘organized as a collection of patriarchally
governed families’. Apart from the biological and comparative ethno-
logical reasons for rejecting such a conception, it leads to a tangle of
difficulties and inconsistencies within the domain of Roman and Greek
archaeology itself. Sir Henry Maine candidly admits that the origin of
the patriarchal family itself is a complete mystery. One of the most
distinguished of classical social students* M. Fustel de Coulanges*
working purely from the classical point of view, felt obliged by Ins
difficulties to break away from current theories, and to regard the origi-
nal unit, not as a family, but as a gens or clan. I have already given some
reasons for thinking that such a group could not have rested upon
patriarchal principles.

In the light of present knowledge, the evidence that Roman society
was preceded by a matriarchal constitution is no less conclusive than
in regard to other people. The primitive Romans were divided into
tribes* and those tribes into curiae, which, as Livy reveals, were named
after each of the Sabine women: that is, they were named after the
mothers and not the fathers of the clan. Thus, the primitive organiza-
tion of the Romans consisted of motherhoods. The Latin people as a
whole derived their name from a tribal ancestress, for the first king of
the Latins was, according to tradition* Satumus and his wife* Latia.
Furthermore, the land upon which Rome itself was built was tradition-
ally inherited from women. Acca Larentia, who is curiously described
as ‘a most noble prostitute’ married a wealthy Etruscan ‘whose home
she ruled,’ and left the land to the Roman people when she died.
Moreover* succession was not in the male line. If a male of the royal
family had a claim to the throne, it was through his mother, and he did
not succeed his father, but his uncle. When Tarquin the Proud caused
one of his nephews to be murdered in the hope of securing the succes-

THE MOTHERS

sion for his own son, the other nephew feigned insanity; and it was upon
him, in accordance with matriarchal law, that the duty of blood-revenge
evolved, when his niece Lucretia was assaulted. As in all patriarchal
society, the distinction between paternal and maternal uncles was
clearly drawn, the former being called ‘patruus’, the latter ‘avunculus’, a
diminutive of avus ; that is, ancestor. Thus our word ‘uncle’, a corrup-
tion of avunculus, preserves a trace of the matriarchal order of succes-
sion.

The Etruscans, the most important of the Italian populations, and
probably the actual founders of Rome, are known to have been defin-
itely matriarchal. There is no instance of an Etruscan agnomen — that
is, of a name derived from the father — and on funeral monumenta the
deceased is usually designated by his metronymic.

In bilingual inscriptions, the father’s name is inserted in the Latin
version only, while the mother’s name, always given in Etruscan, is
sometimes omitted in the Latin. Etruscan girls were unrestricted before
marriage and were said to earn their dowry by prostitution. Their free-
dom after marriage was scarcely less. ‘It is a custom instituted by law
among the Etruscans’, says Theopompos, ‘that wives should be in
common.’ In their frequent feasts the married women lay on couches
with any man they chose and had freedom of intercourse with him.
Paternity, we are told, was unknown. No word for father has yet been
detected in the inscriptions. The words denoting husband and wife
are also somewhat doubtful.

It is difficult to imagine how in primitive times two populations so
closely intermingled as were the Etruscans and the local Italic tribes
could have maintained a separate and totally different form of constitu-
tion, one matriarchal and the other patriarchal. In the bilingual sepul-
chral inscriptions special devices were adopted to render the matri-
archal nomenclature in accordance with later Latin usage. The word
‘agnomen’, the correct term for a patronymic, derived from the paternal
relatives, was only introduced in the fourth century. Theopompos
uses the same language in speaking of the Roman plebs as he does of
the Etruscans: they were said not to know their fathers. But originally
the nobility were in the same case, as Virgil discloses. The most emi-
nent of the Latin nobles, Drances, ‘was proud of the nobility derived
from his mother; as to his father he was uncertain’. This uncertainty
was shared by the kings of Rome themselves — Romulus, Ancus Martius
and Servius Tullius.

The patriarchal principle of transmission of property to the son was
evidently an innovation of the patricians, the owners of property. We
find, in other parts of the world, the propertied classes adopting identi-
cal measures to transmit it in the male line. Thus, the wealthy of the

THE MATRIARCHAL PHASE IN CIVILIZED SOCIETIES 95

Tlinkits of Alaska are patriarchal while the poorer classes are matrilin-
ear. The same is true in Dahomey.

The contest between plebeians and patricians which occupies so great
a place in early Roman history is not merely part of the conflict between
the poor and the rich, but also a conflict between two forms of social
organization. The transition from one to the other appears to have taken
place almost within historical times. The elder Cato refers in pretty clear
terms to the event. ‘Our fathers have willed that women should be in the
power of their fathers, of their brothers, of their husbands.’ Plutarch
represents Roman senators as scandalized at the notion of a woman
raising her voice in the Senate, yet in the age of Romulus women
commonly delivered orations in the Senate. Tacitus notes as a peculi-
arity of the Germans that they insisted upon female hostages, but
Porsenna did exactly the same thing with regard to the archaic Romans.
Much in Roman cult survives from an earlier time. In the temple of
Ceres, for instance, the names of male relatives were never pronounced,
and in the rites of Mater Matuta it was the custom for Roman women
to pray first for their sisters’ children; that is, for the children of the
maternal clan. The Romans, observed one historian, ‘had a most remark-
able predilection for ascribing to women the most important events in
their history.’ Virgil describes the Italic tribes as being led against the
invader by an unwedded queen, Camilla, high-priestess of Diana, who,
though she has a brother living, reigns over the Volsci in her own right.
He pictures the Latin queen, Amata, claiming the right to choose a
husband for her daughter, that is, an heir to the throne, and inciting
the Latin women to resist her husband’s nominee, as their ‘maternal
right’.

It is owing to its original matriarchal character that, in spite of the
‘patria potestas’, women nevertheless retained in that society their
dignity and privileges, which contrast with their position in Greece. As
wc shall see later, their status in Rome was marked by a strange com-
bination of patriarchal institutions and matriarchal sentiment.

To conclude, we see that even in those countries which are now
regarded as strongholds of patriarchy, the former status of women was
higher and more independent than in later times. Possibly the only
exceptions to this rule are the Australian aboriginals and the Melanesian
savages.

CHAPTER 8

Division of Labour

The difficulty which many experience in recognizing the matriarchal
character of primitive societies arises, I believe, from a fundamental mis-
conception. It is assumed that in a matriarchal society the women
dominate the men in a way similar to that in which men dominate
women in a patriarchal society, the difference merely being in the sex
which exercises power. But in fact the relationship is quite different, for in
advanced patriarchal societies men’s power depends upon the fact that
they are producers, whereas women are economically dependent. In primi-
tive society, however, where private property scarcely exists, no such re-
lationship is possible and the conception of authority is not understood.

It is the development of private property and the desire of the male to
possess it which is the commonest cause of the change from matriarchy
to patriarchy, the other frequent motive being the desire for a monopoly
of certain magical powers. The term ‘mother-right’, which is so often
employed in speaking of matriarchies, therefore contains a misleading
implication, for it assumes the notion of established claims, which are
in fact peculiar to patriarchal societies. The primitive ascendancy of
women is founded not on economic power but on the constitution of
the social group. The primitive human group is matriarchal for the same
reasons that the animal group is matriarchal. The group subserves the
maternal instincts and is governed by them, but this fact does not im-
pose a female domination on the male.

Where the matriarchal order exists after the development of private
property, and such wealth continues to remain in the hands of the
women, the result is indeed a female economic domination, similar to
that of men in patriarchies. This state of affairs may be called a gynaeco-
cracy, as distinct from primitive matriarchy.

This does not mean that there are not distinctive economic relations
in the primitive group. The sole form of wealth and power at such levels
consists in power to produce, and the advantage in such power is in
favour of the women, for in primitive society women are the chief
producers. This power is associated with agriculture, which appertains
to women, for the productiveness of the hunter can never go beyond
hand-to-mouth subsistence.

DIVISION OF LABOUR

Women's Labour

It is precisely because the women sustain the main burden of labour
that they are in a position of independence, and it is a fallacy to suppose
that the hard work which they do is an indication of inferiority. No
labour in primitive society is other than voluntary. Numerous accounts
exist of the slavery and oppression of North American squaws ; they rest
upon complete misunderstanding. ‘The work is not only voluntarily but
cheerfully performed’, Father Theodat observes. ‘It is perfectly volun-
tary labour. All labour with Indians is voluntary’, says Schoolcraft.
Furthermore, though the women work hard, the work of the men is
generally far more strenuous and dangerous. The women themselves
say that they have no hardship, for while their labour in the fields em-
ploys them, at most, for six weeks in the year, the men work the whole
year. A single man may have to find food for as many as fifteen people,
lie may have to walk twenty-five or thirty miles over rough ground,
fasting, in order to find game and may have to drag home an 150 lb. deer.

Some travellers have noted that, when savage people make a journey,
the women bear the burdens while the men only carry weapons: but
such an arrangement is essential both for safety and to free the men to
take advantage of any game which may be encountered.

What is true of North American Indian society is true of almost all
primitive societies. Even where women are roughly handled, as in
Australia or Melanesia, this treatment is not used to compel them to do
tasks against their will. The state of such women is by no means so
oppressed as one might imagine. Thus a missionary writes of the Zulus:
‘Whoever has observed the happy appearance of the women at their
work, their gaiety and chatter, their laughter and song . . . let him com-
pare them with the bearing of our own working women.’ And as Shep-
stone says : ‘The labour performed by the women bears no comparison
to what is performed by the women of the lower classes in England. . . .
As a rule, women only work during these eight weeks in the year.’

How did this division of labour between the sexes arise ? Among
mammals there is no economic division of labour except in so far as the
mothers of the young carnivora assist them in procuring food. The same
is true of the higher primates ; in a band of gorillas each animal, male or
female, forages for itself. No doubt, primitive humanity was also mainly
frugivorous, and in those conditions very little division of labour could
take place. As we shall presently see, in primitive society there is very
incomplete division of labour. Women hunted equally with men and
fishing was done more commonly by the women than the men. It was
the development of hunting which finally established the primitive
division of labour between the sexes.

98 THE MOTHERS

This division of labour was not chiefly determined by the respective
aptitudes of the sexes, nor by any physical inferiority in women, but by
woman’s need to care for her offspring, which prevented her from under-
taking pursuits entailing prolonged absence. This handicap is much
greater in mankind than in any animal species, because of the much
longer duration of infancy. An Iroquois myth recognizes this, represent-
ing early men and women as hunting together in the forest, whereas
when children were born there were so many things for the wife to do
that she stayed at home.

Physical Difference between the Sexes

The disparities between men and women in physical power, enterprise
and capacity for endurance, observed in civilized societies, are often
regarded as due to organic differences, but they are much less marked
among primitive men and women. Such differences are often assumed to
be the cause of the division of labour, but seem to be largely a result of it.

Throughout most of the animal kingdom the females are larger and
more powerful than the males. This is true, for instance, of all spiders,
all teleosteous fishes, all snakes, most lizards and many birds, as well as
invertebrates generally. But among mammals the male is almost in-
variably larger. It would appear that there is some correlation between
the physiological development of motherhood, prolonged pregnancy,
etc., and the reduced size of females among mammals. But, though the
mammalian female is generally smaller, there is no sign of physical
inferiority. On the contrary, female carnivores are more formidable than
the males. ‘A lioness’, says Mr Rainsford, ‘is, I think I am safe in saying,
a hundred per cent more dangerous than a lion.’

While women are generally smaller than the men of the same race,
even this is not invariably true. Thus among the Bushmen the women
are, on an average, about four centimetres taller than the men, and
among the Adombies of the Congo the women are reportedly stronger
and better-developed than the men, as also among the Ashira, the
Bashilanga, the Wateita and others. In Dahomey ‘the women are gener-
ally tall, muscular and broad, and the men smooth, full-breasted, round-
limbed and effeminate-looking’. A Kikuyu man is quite unequal to
carrying a load that his women think nothing of. The same is true of the
Manyema women of the Congo, the Krus, the Iyashi of Nigeria, the
Lala tribes and others. Similar reports come from Patagonia, Tierra del
Fuego, Tibet, the Khasi women of Assam, the Gond tribes of Central
India, the Chinwan of Formosa, the Dayaks and others. In Butan, the
women carry the men on their backs when travelling.

Speaking generally, the physical differences between the sexes are
less pronounced in primitive races than among civilized peoples. In

DIVISION OF LABOUR

prehistoric skeletons, the determination of sex is often difficult, since
the bones are as massive in the female as in the male, while the muscular
attachments are nearly as pronounced. Among the Bushmen, it is often
difficult to distinguish the sexes in life, even where the individuals are
almost naked. (The breasts constitute no distinction. They are so
developed in the males that these are sometimes able to suckle.) Similar
difficulties are reported by Phillips of the Lower Congo races, and of the
Eskimo and other peoples.

Women as Hunters and Fishers and as Warriors

That men took on the labour of hunting was not due to incapacity on the
part of the women. In British Columbia the women used to hunt ; the
women of New Spain and of Nicaragua hunted regularly; the same is
reported from Tierra del Fuego, Tibet, West Africa, Tasmania and else-
where. Procopius mentions certain barbarians of Thule among whom
men and women hunted together.

Fishing is often exclusively a women’s occupation (as for instance
among the Bambala, the Tasmanians, the Fuegians), and in some seasons
the women provide the whole food supply of a tribe.

In short, the primitive division of labour has become established more
by a spirit of professional exclusiveness than by a difference in aptitude.

Similarly the tendency of the man to take on the role of warrior and
fighter is also due to economic necessity, rather than to any constitutional
incapacity of women. In New Guinea, Malaya and many parts of the
Pacific, including New Zealand and Hawaii, women were accustomed
to fight alongside the men, and in the Ladrones they fought under
female leaders. As late as 1900, when punitive expeditions were sent
against an inland tribe in Papua, the police found that the houses were
garrisoned by as many women as men. Women frequently accompanied
the men to war among the North American Indians; the title of ‘Be-
loved’ was bestowed upon those who had distinguished themselves by
warlike deeds. During their expedition on the Klamath in 1854, the
us troops frequently saw women fighting or found them dead. ‘One day
the savages came suddenly upon them . . . filling the air with a perfect
shower of arrows. But not a male barbarian was in sight. Before them,
in serried line of battle, their women were moving to the charge, while
the warriors slunk behind them, discharging their arrows between the
women.’ In the province of Cartagena the Spaniards found that ‘the
women fight as well as the men’. Columbus and his companions were
attacked by female archers.

In many Arab countries the women fight with the men; the Sultan of
Zanzibar had an army of six thousand female soldiers. These were
certainly not ornamental. In the case of the famous corps of Amazons

100

THE MOTHERS

of Dahomey, to whom King Gueso owed his safety, the women alone
stood their ground, in spite of appalling losses, while the rest of the
army scattered. Their strenuous training involves taking fortified posi-
tions by assault, charging through obstacles formed by cactus and spikes
and, as they march past after these sham fights, their bodies stream with
blood and the skin hangs in shreds from their torn limbs.

The Tartar and Mongol women of Central Asia have long been noted
for their war-like qualities; Matthew Paris, describing the Mongol
hordes of Jinghis Khan, refers to the fierceness of their wives who
‘fight like the men’. The kings of Siam used to keep an Amazonian
guard, as did the Nizam of Hyderabad; the kings of Kandy kept a body-
guard of female archers, as did the Persian kings. Of the ancient Scyth-
ians, Diodorus reports that ‘the women fight like the men and are in no
wise inferior to them in bravery’. There is therefore no reason to dismiss
the Greek accounts of their battles with Asiatic Amazons.

Celtic women regularly followed their male relatives in war, and such
military service was not abolished until 590. Irish literature abounds
with references to war-like women. Tradition preserves the memory of
chieftainesses such as Geraldine Desmond who ‘was of a fierce and
restless character’, and led her clansmen constantly into frays, killing
all who opposed her.

When the Romans had their first terrible encounter with the over-
flowing barbarian tide at Aquae Sextiae, ‘the fight had been no less
fierce with the women than with the men themselves’. When the bar-
barian onslaught broke, the women erected barricades and fought
fiercely amidst the mountains of dead ; when summoned to surrender,
they killed their children, slaughtered one another and hanged them-
selves on trees. Among the slain Marcomanni and Quadi, the Romans
found the bodies of women in full armour. Several of the Gothic war-
riors taken prisoner turned out to be women, and the same was ob-
served by the Byzantines when they were attacked by the Varangians.
Queen Boudicca boasted that British women were quite as good war-
riors as the men.

Of the endurance and courage of primitive women, an even more
general test is constituted by their behaviour in childbirth. Though child-
birth appears to be, as a rule, easier with primitive than with civilized
women, nevertheless their endurance is remarkable, and in many
tribes if a woman uttered a cry during labour she would be disgraced.

Primitive Industry

Thus it is clear that the differentiation of men as hunters was not due
to the more limited powers of endurance of women. At the same time,
the industrial occupations which give rise to material culture belong in

DIVISION OF LABOUR IOI

the rudest societies almost exclusively to the women. All industries
were at first home industries, and developed therefore in the hands of
the women.

The preparation of skin as apparel was a function of women, and was
carried to high levels of skill. The scrapers used by Eskimo women
today for preparing skins are identical with those found in the drift
gravels of the Ice Age, and in South Africa the country is strewn with
scrapers, manufactured by the women of the Bush tribes, which are
identical with those of Palaeolithic Europe. The flint arrow-heads of the
Seri Indians, which are identical in type with those of Palaeolithic
Europe, are said to be manufactured by the women.

The decorations on a fur robe and the combination of various furs
may be so elaborate that it may take a whole year to complete a single
garment. The sewing and ornamenting of robes even form the chief
purpose of a kind of 'secret society’ among North American Indians;
a society from which men were excluded.

A similar development occurs in those areas where women concen-
trate on weaving or plaiting. Thus in New Zealand, where there are at
least twelve different styles of mats differing in the fashion of plaiting,
each is the speciality of the women of a particular tribe. Initiation into
the art was conducted as a religious ceremony; a consecrated workshop
was reserved for the purpose, and if a man entered the precincts all
work was stopped and put aside. Similarly, among the Manipur tribes,
cloth in ten different patterns is produced in certain villages only, this
industry being carried on by the women alone.

Pottery

The art of pottery is likewise found in the hands of women among all
primitive peoples. Only under the influence of advanced culture does
it become a man’s occupation. Thus, out of seventy-eight tribes investi-
gated by the ethnologists attached to the Belgian Congo Museum,
‘the men had no hand whatever in the making of pottery in sixty-seven’;
the others ‘are exceptions arising from special circumstances which in
almost every instance it is easy to trace’. If we follow the course of the
Nile we find a complete series, from Nubia where, at the present day,
the pottery is made exclusively by the women, through Upper Egypt,
where the head-potter is a man with women working under him, to
Lower Egypt where the pottery is made by the men. And there is no
reasonable doubt that prehistoric pottery was the work of women.
Remains found in the lacustrine dwellings of Switzerland bear numer-
ous imprints of thumbs and fingers ; undoubtedly those of women.

In this field, too, the art becomes a mystery, the secrets being trans-
mitted in the female line. Every Samoki woman must be proficient in

102

THE MOTHERS

the art, but she is forbidden to practise it if she marries in another
district. The same thing is noted in Ceram and among the Massims
of New Guinea. Pottery is unknown in most parts of Melanesia, but
in those where it exists it is likewise a monopoly handed down from
mother to daughter. In East Africa, among the Nandi, no man may
go near the hut where women are engaged in making pottery, and if a
man should take a woman’s pot he would be sure to die. In Manipur if a
man approaches while a woman is making pots they will crack in the
firing; as also in Brazil among the Tupis.

The Greek tradition that the first ‘patera’ w r as moulded on the breast
of Helen is illustrated by the practice of the Zuhi women who, as of
old, make their pitchers in the shape of a female breast. The nipple is
left open to the last and the sealing of it is performed with the solemnity
of a religious rite. The vessel is thus assimilated to the woman herself,
for unless this ritual were observed that woman might be barren. Such
an assimilation is widespread. In the earliest Egyptian hieroglyphics
the pot is the symbol of womanhood, and vases used among the archaic
populations of the ALgean were frequently in the shape of a woman.
In Greece the pot was an emblem of fecundity. In Cyprus, at Lithro-
donto, a potter of today puts two little dabs of moist clay on the right
and left side of the rounded surface, slightly above the middle, little
knowing why he does so. But vases disinterred by archaeologists from
tombs 2,500 years old display numerous specimens with the same
finishing touch — the two dabs representing two breasts. ‘The Mother
Pot is really a fundamental conception in all religions,’ observes Dr
Elliott Smith, ‘and is almost world-wide in its distribution.’

The sacred vessels used in religious cults often retain their archaic
character. Thus the pots used by the Roman vestals were, even in the
height of Roman splendour, made in the coarsest fashion without the
use of a wheel, and were spoken of as ‘Numa’s crockery’. There can be
little doubt that in early times the potters were the vestals themselves.
In the same manner the vestals of Peru served their God in very coarse
vessels of clay, until the Incas, shocked at this, had them replaced by
vessels of gold. The Zuhi women, with an aesthetic sense that would
have rejoiced Ruskin, scorn the wheel — a comparatively late develop-
ment — as giving a machine-made appearance to the products, and trust
entirely to the eye. 1

Thus we may conclude that the art of pottery was a feminine in-
vention.

1 The patterns with which the clay is ornamented are commonly derived from
the braiding of basket-work, and early pots were no doubt shaped in or upon
baskets. Much contemporary pottery preserves these patterns, and the ‘Greek
fret’ need be no more than a basketry version of a ‘loop-coil’ meander.

DIVISION OF LABOUR

103

Building

We are not accustomed to think of architecture as a feminine occupation.
Yet, just as the animal builds its nest or burrow, so also primitive
woman was the actual home-builder. The huts of the Australian, of
the Andaman islanders, of the Patagonians, of the Botocudos, the rough
shelters of the Seri, the skin lodges and wigwams of the American Indian,
the black camel-hair tent of the Bedouin, the ‘yurts’ of the nomads of
Central Asia, arc all the exclusive work of women. But even the most
elaborate buildings of the uncultured world were fashioned by women.
The earth-lodges of the Omahas were built entirely by them, as were
the elaborate pueblos of New Mexico and Arizona, with their court-
yards and piazzas. Among the Zuni today the men assist with the
heavier work of timbering; but among the Hopis the work is still done
entirely by the women. When the first Spanish priests settled among the
Pueblo Indians, no man had ever set his hand to the erection of a house.
In the building of the larger and more substantial timber houses most
of the heavy work is generally undertaken by the men, but the tradition
of the primitive division of labour survives in the fact that the men are
invariably assisted by the women. In Samoa, for instance, the women do
the thatching, though the men erect the framework. In Egypt even in
the present day the building is done by the women, and the assistants
of the master builder arc invariably girls.

It is a significant detail that, in most parts of the uncultured world,
the maximum of care is bestowed on buildings in which food is stored,
and which therefore belong to the province of women. Sir Charles
Fellows was one of the first to remark that the food stores in Lykia and
other parts of Asia Minor at the present day are the exact counterpart
of the ancient tombs and temples.

The primitive temples of Greece were built of wood. The earliest
temple of Delphi was a thatched hut. In Lydia, the dwelling-houses, as
Herodotus mentions, were mostly built of reeds and mud. There can
be little doubt that, like the African kraals which they resembled, they
were built by the women. The most imposing monument in the country,
the heroon of Alyattes, bore on one of its columns an inscription stating
that, for the most part, it had been built by women.

Primitive Trade

Since all surplus production belongs to the women, both as cultivators
of the soil and as keepers of the food store, it is theirs to dispose of. In
all early culture, the barter and traffic is in the hands of the women.
This is true throughout Africa, where trade is almost exclusively
carried on by the women. Formerly the North American fur trade was

THE MOTHERS

104

in the hands of the women who prepared the skins. In Nicaragua ‘a
man may not enter the market or even see the proceedings, at the risk
of a beating’. Throughout Central Asia trading is entirely in the hands
of the women, as Marco Polo also reported. This is also true of Tibet,
Burma and many parts of the Pacific.

Medicine and Surgery

The search of women for edible vegetables acquainted them with the
properties of herbs and made them the first doctors. The word ‘medi-
cine’, and the name Medea, the medical herbalist witch, come from the
same root— a root meaning knowledge or wisdom.

Primitive medicine is of course for the most part a department of
magic, and women, as we shall see, were the primitive practitioners of
this art. In addition, in many areas the women arc the surgeons and treat
all injuries received in warfare. Thus, throughout the primitive popu-
lations of Indonesia, the treatment of the sick is almost exclusively in
the hands of medicine-women. Among the Land Dayaks, while there
are both male and female doctors, the service of women is regarded
as more valuable and is more highly remunerated. Among many North
American tribes, as also among the Eskimos, the profession is practised
solely by the old women. The same is true in Patagonia, Chile, Mexico
and elsewhere.

In early Arabia and down to the Middle Ages the treatment of wounds
and the practice of medicine were recognized occupations of women,
and even at the present day this tradition survives in the more secluded
Muslim countries. Among the ancient Germans, Scandinavians and
the inhabitants of Gaul, women were regarded as especially qualified
in medicine.

In short, to a far greater extent than is generally realized, the sexual
differences between men and women are products of social develop-
ment. Primitive woman is anything but ‘effeminate’, and is as capable
as the man of providing for herself, if she is not indeed his superior.

It is commonly supposed that the facts of primitive matriarchy favour
the doctrine of feminism — that is, the view that women are fitted for
all the pursuits which in our society are regarded as masculine. This
view, whether correct or not, docs not follow from the respective
capacities of men and women in primitive society. The predominance
of women in primitive society, which appears to some so paradoxical,
would be largely restored in our own society if culture were limited to
the range of activities of the primitive household.

In the spheres which are important at this cultural level, intellectual
advantage is with the female. She is more wary and ingenious than the
male, and it is little wonder that the savage habitually goes to his women-

DIVISION OF LABOUR IO5

folk for advice. In many societies it has been remarked that the women
are mentally higher than the men, and they are far more self-possessed.
For instance, according to Mr Landor, ‘the Tibetan woman is far
superior to the Tibetan man. She possesses a better heart, more pluck
and a finer character than he does. Time after time, when the male,
timid beyond description, ran away at our approach, the women re-
mained in charge of the tents and, although by no means cool and
collected, they very rarely failed to meet us without a show of dignity.’
The same thing may be observed in the ruder strata of western society :
the French peasant woman, for instance, is a more intelligent, alert and
less awkward person than her man.

Position of Chiefs

It has sometimes been thought that the fact that in matriarchal societies
the headman or chief is commonly a man is an argument against the
view that primitive society was essentially matriarchal. This is due to a
misunderstanding: the position of the headman in primitive societies
entailed nothing of authority, for there is no such authority in truly
primitive societies. Those whom the whites call headmen are usually
little more than spokesmen, put forward to voice the views of the group
and without any power to act on their own initiative. In Northern
Melanesia, the transacting of any business with the natives is ‘rendered
very difficult by the fact that they have no chiefs’, while in New Cale-
donia the chiefs have ‘absolutely no political power’. Among the tribes
of Assam, ‘each village is a small republic and each man as good as his
neighbour; indeed, it would be hard to find anywhere else more
thoroughly democratic communities. Headmen do exist, but their
authority is very small.’

Such tribes have often found the necessity of appointing a leader for
hunting or war. Thus a missionary reports of the Indians of Brazil, that
‘they know neither princes nor kings; each family regards itself as
absolutely free ; every Indian looks upon himself as independent. As the
continual wars which they have to wage against their neighbours place
that liberty in danger, they have learnt the necessity of forming a sort of
society, and they choose a chief who is called “cacique”. But in choosing
him, their intention is not to give themselves a master but a protector
and father under whose guidance they desire to place themselves.’
Similarly, among the Iroquois, the chiefs are nothing more than the
most respected among their equals in rank. Their principal duty is to
conduct negotiations, and if they make a mistake they are reprimanded
or dismissed.

Among the Arabs, the chief is merely ‘influential’, and is obliged at
every turn to consult the tribal council. Even in Africa, the land of

T.M. — 4*

10 6 THE MOTHERS

despots, the chiefs are not always what the European assumes. Often
they are no more than war-leaders. Thus Dundas states, of a number of
tribes of East Africa : ‘After the most careful inquiry ... I feel convinced
that these tribes have no heads or leaders who could be dignified with
the names of chiefs.’

Thus the existence of male chiefs in primitive social groups, far from
constituting a difficulty as regards their matriarchal nature, appears, on
the contrary, to prove the impossibility that such groups could ever have
been formed in the first instance around the authority of a dominant
male. Such domination is incompatible with the equalitarian character
of primitive societies.

The desire to dominate takes different forms in men and women.
Woman’s ambition is not to exercise physical compulsion, but to bend
the will of others to her desire, to overcome not physical but psychical
resistances. Her object is to have her way. The man, however, delights
in the display of power, is flattered by hostile envy and desires his power
to be felt and known. Thus it is that, where tribes have conquered other
peoples, the war-leader has acquired absolute authority over them.

The point is further emphasized by the fact that, in matriarchal
communities, the authority of the male headman is even more insigni-
ficant than in other primitive communities. Thus among the Seri the
chief appears to do little more than communicate the decisions of the
matrons to the men ; he is not chosen with regard to his own qualifications
but to those of his wife. In Khyrim, the chieftainship is limited to the
male relatives of the high -priestess. Among the Pelew Islanders the
authority of male chiefs is exercised over the men only ; women do not
even salute them, and they can take no action without consulting the
council of matrons.

But even when the theory of male rule has become fully developed, the
authority of the woman is completely recognized within the home and
the economy of the family. The primitive division of functions has some-
times left its trace in comparatively advanced communities. In Hawaii
the administration of all internal affairs was under the control of the
queen, the king being concerned with foreign politics only.

Position of the Woman? s Brother

In his character of warrior, the man is defender of the home, and in
him is vested the executive power of action in the relation between the
home and a hostile environment. But the male to whom a woman looks
for this protection is not her sexual mate but her brother. The word
brother, in Sanskrit ‘bhratr’, comes from the root ‘bhr’, to support.

The position of the eldest brother of the woman is so well-known a
feature of primitive society that it need not be described in detail. It has

DIVISION OF LABOUR

107

frequently survived the original constitution. Thus, in North America
the relationship of uncles is, in several particulars, more important than
any other. In every part of the African continent the mother’s brother is
responsible for his sister’s children. For instance, on the Gold Coast a
man may chastise or even sell his sister’s children, but in no circum-
stances may he do so with his own, since they do not belong to his
family but to the mother’s. Among the Adiokru of the Ivory Coast
province, legitimate children are under the care of their father, illegiti-
mate children (who suffer no social disability) under the care of their
mother’s brother — the matriarchal and patriarchal customs thus existing
side by side. Among the natives of Timorlaut, although the social organ-
ization is now essentially patriarchal, the mother’s brother occupies a
fundamental position and performs most of the parental functions.

We have already noticed this relationship among the ancient Germans,
and we have found distinct traces of it in primitive Rome. Among the
Christian Ossetes of the Caucasus, before a man can marry he must
present a horse to the bride’s maternal uncle. Today no social organiza-
tion is more definitely patriarchal than theirs. In Krete, in the seventh
century bc, when patriarchal usages had become well-established, a
woman’s illegitimate children were brought up by her brother.

The brother is also the partner in the economic association which
constitutes the primitive group. Moreover, the main object of a woman’s
affection is her brother rather than her sexual mate. If we have grasped
the fact that the extension of affection towards the male has the sole
object of enlisting his co-operation, we shall have no difficulty in under-
standing why a woman’s affection for her brother may be as lively as
that for her husband — if not indeed more real, for sexual love always
contains a hidden element of opposition. In primitive societies, the
bonds between brother and sister are generally stronger than those
between husband and wife, and the existence of the latter is often
doubted. Thus among the Hovas of Madagascar, where love between
husband and wife is said to be ‘not even thought of’, there is ‘no lack of
affection between brothers and sisters’. The closest relation existed
between brother and sister among the North American Indians: if a
young warrior captured a horse it was to his sister that he presented it;
it was invariably his sister who advanced to meet him when he returned
from battle. Even where traditional usages of avoidance place a barrier
between brother and sister, as in New Caledonia, they remain extremely
fond of each other. In Samoa, the relationship is regarded as ‘semi-
sacred’.

In Sophocles’ Antigone , the heroine sacrifices herself not for her lover,
who plays a subordinate part in the drama, but for her brother, and
expresses emphatically her reasons for the preference. This passage

io8

THE MOTHERS

horrified Goethe, but exactly the same sentiments are expressed in the
oldest Teutonic literature. Herodotus also relates how the wife of Inta-
phernes, when Darius gave her the boon of asking for the life of one of
her kinsmen, chose her brother, and this so pleased Darius that he gave
her the life of her eldest son also.

A widely current Arab proverb reproduces this reasoning. ‘A husband
can be found, a son can be borne, but a brother cannot be replaced.’

The Marital Group Family

Prescientific theorists have generally regarded the family as the original
unit of primitive society, for as an eighteenth-century writer elegantly
puts it: ‘The husband and the wife of his bosom, whom love unites by
the silken ties of matrimony, form the first society; this union is first
founded on the call of nature, in mutual assistance, and the sweet hopes
of seeing themselves reproduced in numerous offspring.’ The facts of
ethnology do not accord with this dogma. Primitive social organization,
on the contrary, is based on common blood and on derivation from the
same womb — the so-called uterine family. Thus among the Fanti, each
family includes members of the mother’s side only ; the mother and all
her children, male and female belong to her family, as do her mother
and maternal uncles and aunts, but her father and all his relatives are
nothing to her nor are her husband nor any of his relatives.

In many primitive societies there is no word for the concept of family,
though there may be a word which indicates marriage. Thus Mr Joel-
son says, speaking of the natives of East Africa generally: ‘For want of a
better word, I must needs refer to the Negro “family”, but my readers
will realise that the term in this connection is used to convey an idea
essentially different from the construction put upon it in modern society.’
Of the ‘family’ in West Africa, Clozel said: ‘Although apparently it
resembles that of European societies with father, mother and children,
the father’s authority scarcely exists and from the civil point of view he
is not the parent of his children. The true family among the Alladian
only takes account of the uterine parentage’. Among the Wyandots, says
Mr Powell, ‘the family household is not a unit, as two gentes are repre-
sented in each, the father must belong to one gens, the mother and her
children to another’. ‘The Indians consider their wives as strangers’, says
an old missionary. ‘It is a common saying among them, “My wife is
not my friend”, that is “she is not related to me and I need not care for
her”.’

The husband was isolated; plans and secrets existed among the mem-
bers of the gens rather than between husband and wife. The children
do not regard their father as a relative by blood; if he requires assistance
they consider that ‘his people’ should look after him. In disputes, the

DIVISION OF LABOUR 109

children and the father in primitive society fought on opposite sides.
Among the Haidas ‘it appears as if marriage were an alliance between
opposite tribes, a man begetting offspring rather for his wife than for
himself, and being inclined to see his real descendants rather in his
sister’s children than in his own. Husbands and wives did not hesitate
to betray each other to death in the interests of their own families’.
Among the Goajiros, we are told again in so many words, ‘what we call
the family does not exist’. Numerous other cases can be instanced.

Not only docs the family not exist as a juridic or social unit, it fre-
quently does not even comprise a physical association. It is common in
primitive society for husband and wife to live apart. In New Caledonia
‘domestic life does not exist’. Men and women are seldom seen talking
or sitting together. The men generally associate with other men and live
in their club-house. The Nayar husband is not even permitted to take
food in his wife’s house; indeed, it is a universal rule in India that men
and women do not eat together. There is no common life between hus-
band and wife in China ; the house is divided into two and the sexes live
in separate compartments. In Korea ‘family life as we have it is quite
unknown’. The same is true of the Eskimo. There was scarcely any
intercourse between men and women in any of the North American
tribes; a man did not take his meals with his wife, and rarely spoke to
her in public. So fundamental was the custom that it was kept up by the
completely Christianized and semi-Europeanized Indians living on the
outskirts of Quebec. Similar customs can be found among the Pueblo
Indians, the Caribbean tribes, the Carajas of Brazil and in many parts of
Africa.

Even more extraordinary in our eyes is the fact that the husband is
commonly a clandestine and surreptitious visitor to his wife. One of the
Japanese words for marriage is ‘yome-iri’, which may be interpreted
‘to slip by night into the house’, and this is the common mode of con-
nubial intercourse among many primitive peoples. Among the Khasis
‘the husband came to his mother-in-law’s home only after dark, and he
did not eat, smoke, or even take betel-nut there’. Among the Tipperah,
the Takut, the Samoyeds, the Kurils, the Tartars and other tribes, the
husband slips surreptitiously to his wife’s room after dark and leaves
before dawn. If a Tartar is seen leaving, he receives a sound drubbing
from his wife’s relatives. In Khorassan, it was the rule for the mother of
the bride to introduce the bridegroom secretly into the house by the back
door, and he had to depart before dawn. Among the Cherkiss the hus-
band gets in through the window; much the same is true throughout
the Caucasus.

The same practice was found in the primitive community of Sparta.
It is reported from Africa, and of the Algonkin and Iroquois in North

THE MOTHERS

America, the Caribs of the West Indies and many tribes of New Guinea,
as well as those of Fiji and elsewhere.

The idea that this behaviour was due to some innate sense of delicacy,
though met with in apologetic literature, cannot be taken seriously, for
those natives who would be filled with confusion if seen approaching
the hut of their wife have at the same time no scruple against indulging
in promiscuity, and even incest, coram populo. There is good reason to
suppose that the relations between husband and wife were attended
with real danger. The beating which the Tartar husband incurs, if
detected, is more than a ritual ; and similarly in Kamchatka it is ‘beyond
a joke’. A Russian story describes how a suitor, when he finally obtained
his bride, was unfortunately rendered impotent by the barbarous treat-
ment he received. Among the Mongols not only the bridegroom but all
his male friends were thrashed by the dowagers of the bride’s family.
The same sort of thing, it would appear, happened in ancient Arabia.
Among certain Germanic tribes patrilocal wedding appears to have
been somewhat risky; the ancient laws of Frisia provided for the bride-
groom being murdered during the ceremony — the bride was instructed
to follow his corpse home if she wished to claim her share of the inheri-
tance. Many of the customs which have been described as survivals of
‘marriage by capture’ are in reality manifestations of this general hos-
tility between the wives’ and husbands’ group, especially when it is
required that the woman should leave her group and join her husband’s
tribe. Swanton goes so far as to say that among the Haidas the members
of the two clans regarded each other as downright enemies.

Accordingly it is difficult to believe that the family as we know it was
the germ of social organization, and that society was formed by the
aggregation of family groups. Indeed, it is dear that the patriarchal
family group is actually in direct conflict with the instincts which have
determined human association. The basic group which the patriarchal
group has superseded is the Motherhood — that is, the mother and her
offspring. It follows that the primitive social instincts of society are not
the sexual instincts but the maternal instincts and the ties of kinship
that derive from them. To these forces sexual instincts are secondary.
‘Love of the clan’, says an old Arab poet, ‘is greater than the love
between husband and wife.’

CHAPTER 9

The Institution of Marriage

What is the origin of marriage ? It is sometimes said that the purpose
of marriage is the procreation of children, but evidently marriage is not
essential to procreation. Among animals, mating is far rarer than is
supposed. There is no evidence of it among the anthropoid apes, and
among those mammals which do mate it usually takes the form of an
association which nearly all human races would regard as incestuous.
Although certain birds and mammals gain economic advantages by
associating, this is not so in maternal clans, where these advantages are
already secured by the female’s association with brothers or with other
males of the group who are not sexual partners.

In some communities only a small minority of the people enter into
marriage, even though the marriage institutions are themselves fairly
elaborate. For example, among the Line Islanders marriage is a method
for conveying landed property, and only the property-owning class
marries. In the Society Islands marriage seems to have been confined to
the higher class of chiefs; while in the Pageh Islands marriage is some-
times contracted by elderly men who wish to provide a home for their
old age.

Even those who seriously believe that husband and wife constitute
the primary social unit recognize that marriage is a social institution and
not an instinctive necessity. Most peoples in all stages of culture have
held this view. The Roman jurists were quite clear on the point. They
made a basic distinction between natural and social or juridic facts.
Natural law, they said, is what nature has taught all animals; it is not a
law peculiar to the human race. The maternal relation, they considered,
was a natural fact, but the paternal relationship belonged to the sphere
of juridic institutions. Again, matrimony is the name of a natural fact,
for it derives from the relation of motherhood; marriage, however, is
the name of a civil institution.

In many traditions, marriage is stated to be a relatively late institution.
The Greeks believed that marriage had been instituted by Kekrops, and
that previously promiscuous cohabitation had been the custom. The
Chinese thought similarly, substituting Fu-Hi for Kekrops, while the
Peruvians attributed the establishment of marriage to Manco-Capac.

112 THE MOTHERS

Even such uncultured people as the Lapps, the Wogul and the Austra-
lian savages attribute the invention of marriage to some traditional
legislator.

If there were really any ‘mating instinct’ which determined marriage,
such an instinct should, in primitive society at least, take precedence
over all other considerations. The personal desire of a man and woman
to enter into a fundamental association should be the basis of the social
order ; but in point of fact the exact reverse is the case. It is regarded
almost everywhere as irregular that a man and a woman should spon-
taneously form such an association on their own initiative. Such a thing
would have shocked our own grandmothers ; and the Australian Black-
fellows, like most other savages, would have shared their feelings.
Marriage, on the contrary, is a matter arranged by the families : among
the Hidatsa Indians a marriage not so arranged is scarcely looked upon
as deserving the name of marriage ; indeed, they do not call it a marriage,
but have a special term for so disreputable a relationship. The same is
true among the Haidas ; a West African negro explained in court that a
certain person would hardly be regarded as better than a bastard since
‘his parents married for love’. Among the Malays of the Patini States, a
runaway match is not regarded as legal. This is, in fact, almost univers-
ally true in primitive society. Even in England, where personal freedom
in such matters has always been greater than in most other countries,
up to the eighteenth century a marriage without parental consent was
null and void when one of the parties was under the age of twenty-one,
and the position is similar in most other European countries today.

In Russia, ‘among the upper classes the bride and bridegroom never
saw each other before the wedding; among the people they saw each
other but never dared to speak about marriage, that being a thing which
did not depend upon themselves’. Among the Romans marriage was
purely a family contract, and in ancient Athens, ‘we have not a single
instance of a man having loved a free-born woman and marrying her
from affection’. The Spartans, who represent a more primitive state of
Greek society, ‘considered marriage not as a private relation . . . but as a
public institution’. The same views obtained among the barbarians of
Northern Europe. The laws of the Aryas expressly condemn as immoral
‘the voluntary union of a maid and her lover’. Such unions are ‘blame-
able marriages’.

In China the most fundamental principle of the elaborate marriage
institution is that marriage is an alliance between two families, and that
the bride and bridegroom have no concern in it — so much so that it is
not uncommon for two families to be united by matrimonial alliance,
even though one, or both, of the parties concerned is dead. Further-
more, the negotiations must be conducted by a professional go-between.

THE INSTITUTION OE MARRIAGE 113

and the young people first make each other’s acquaintance when the
transaction is completed. Any infringement of this rule is looked upon as
profoundly immoral. ‘When children do not await the decision of their
fathers and mothers and the words of the go-betweens, but bore holes
in walls in order to see each other, leap over walls in order to meet, their
fathers and mothers and all the people of the realm despise and con-
demn them.’

Marriages negotiated by Go-betweens

Go-betweens are, indeed, used widely; in Australia, for example; in
Borneo, among the Dayaks; in the Ladrones, the Banks and Loyalty
Islands ; in Africa among the Basutos and Bambara, as well as a number
of other African peoples. They are an indispensable institution among
the Arabs, as well as among many South American tribes. Among the
North American Indians, ‘courtship is always begun by proxy’. Go-
betweens are found in much the same forms in Tibet, upper Burma and
among the Tamil populations of Ceylon, as in Cambodia, Cochin-
China, Siam, Java, Formosa and the Philippines. In Europe the same
is true among the Poles, the various Slav races and the Finns.

In many societies the marriage is arranged when the parties con-
cerned are still infants, a practice by no means confined to sophisticated
aristocracies. The allotting of girls at birth, or before, is the universal
rule among the Australian aborigines . 1 The usage is prevalent, though
not universal, throughout Melanesia, and was common in the case of
men of distinction among the Polynesians . 2 3 It is general in New Guinea,
and the children arc commonly married and cohabit when hardly able
to toddle. :i It is found in every part of Micronesia and the Philippines,
although there it is an occasional means of cementing an alliance be-
tween families and not a general practice. It is common in Indonesia, it
is very prevalent among all the peoples of Northern Asia, and appears
to be almost universal among the Turkic populations. As is well-known,
it is an obligatory practice among the Hindus, and appears to be an
original usage with the Tamil populations. It is the rule among the
primitive forest tribes of Malaya; and in Africa, where marriage by
purchase has reached its highest development, that purchasing power is
extensively used to secure girls in infancy. Though prevalent among the
Eskimo and Alaskan tribes, it is comparatively rare among the North
American races of the eastern regions and the plains. Infant betrothal is
also general in South America.

1 A bibliography of forty-two titles is appended in support of these and the
following assertions in the unabridged edition.

2 A bibliography of forty- three titles is appended.

3 A bibliography of fifteen titles is appended.

THE MOTHERS

114

Where patriarchy is the rule, it is almost invariably the father who
has the right to dispose of his daughters, as for example among the
Kaffirs of South Africa. Among the Papuans of Geelvink Bay, the father
obtains a wife for his son while the latter is still an infant. Sometimes
the selection of a wife is left to a chief or king, as among the Bantu. In
matriarchal societies and those which have preserved matriarchal tradi-
tions, the duty often devolves upon the maternal uncle, and it is his
consent which is essential among the North American Indians. This
idea is so inherent in Indian thought that a current proverb asks: ‘If a
girl is not given in marriage by her mother’s brother, then who will give
her ?’ We have already noticed the same fact in many parts of Africa.
Elsewhere, however, it is the mothers of the respective families who
arrange the marriages, as among the Western Eskimo. Among the
North American Indians, ‘the first proceedings must be initiated by the
matrons’. (Though it is the father who gives his consent.)

Personal Inclinations

Some writers think this custom tyrannous; others note that the personal
wishes of the parties are often deferred to. These opposed points of
view rest upon a misconception — upon the fallacy of attributing to
primitive humanity sentiments which are really peculiar to advanced
cultures. The majority of those concerned have no personal preferences
in the selection of partners and, in general, there is no objection to the
selection made by the authorities. Where objection is made it is often
deferred to, and frequently the objection is only to the partner being
too old, not an objection to his particular personality. The selection of a
wife is often undertaken by the authorities at the young man’s request ;
it is a wife he wants, not a particular young woman. Thus among the
Basutos, when a young man wishes to marry, all he does is to drive the
cows out of the kraal and let the calves be suckled by their mothers. This
conventional sign is understood, and the young man’s parents immedi-
ately set about procuring a wife for him. A Catholic missionary among
the Dene relates how the Indians often consulted him about a proposed
match, and he at first endeavoured to give them his best advice as to the
pros and cons of the alliance. He soon discovered that what they wanted
was not advice but a peremptory order to marry. Often, too, he would
be told by those whose marriage had turned out unhappy that they were
perfectly aware when they married that they could not agree, but they
were told to marry and therefore had no option. Similarly, among the
Bataks of Sumatra girls readily refuse suitors possessing every attrac-
tion in order to marry the young man to whom they are assigned by
custom. If questioned, they say: ‘It is the custom. What else is one to
do ?’

THE INSTITUTION OF MARRIAGE 115

So universal a practice could not have arisen or have been maintained
in equalitarian primitive societies had there been any real antagonism
between such actions and the individual desires of each member. It is
easy to show that in such contracts the wishes of individuals are very
seldom set aside. In West Africa, for example, should a girl refuse to
fulfil the contract, ‘the tendency is not to bring compulsion to bear on
the woman but to enforce restitution of the dowry paid to the girl’s
father’. Indeed, the deference paid to the personal objections of the
young man or woman is much more marked in primitive societies than
in most civilized communities where prearranged marriages are the rule.
The individualistic interests aroused by the profitable disposal of mar-
riageable daughters may, on the other hand, lead to the use of coercion,
and an incompatibility which did not originally exist between the wishes
of the individual and the authority of the group thus develops. The
objections then raised are a sign of the decay of old customs. In the state
of society in which such customs originated, there was no conflict
between the individual and the tribal will.

Marriages Celebrated among all Members of Two Groups

The authority in such matters derives not from the father but from the
group, and defiance of the group’s authority is the most heinous of
crimes; it is usually punishable by death. But in its original form the
group’s control refers not to unions between individuals but between
groups of individuals. A person is bound to marry 7 within a certain
group, but within the group freedom of choice is, as a rule, not inter-
fered with. For primitive legislation takes little account of private
affairs, and concerns itself exclusively with matters which affect the
whole group. Thus, among the Arawak a man is bound to marry — and
is, in fact, allotted at birth to — a certain group of females. To marry
outside that group would be inconceivably impious, but within it he is
free to choose as he pleases. Thus it is that stringent determinism may
be combined with the greatest freedom of personal inclination. Count-
less primitive peoples are in the same situation. Thus among the Gonds,
if a brother and a sister have respectively a son and a daughter, it is
compulsory for these to marry, but if a girl has several cousins she has
the right to choose among them.

In those societies where the father, maternal uncle or mother dispose
of the young person, they do so not in their personal capacity but as
representatives of the family; it is true that, especially in patriarchal
society, this may develop into a personal despotism, but this is a later
abuse. The will of the head of the group is rarely exercised without
consultation with all its members. Thus when the Malaiali father sets
out to find a bride for his son he is accompanied by his relatives, and the

ii 6

THE MOTHERS

first person he approaches is neither the intended bride or her parents
but the headman of a neighbouring village. A Kadir young man must
work in some village for a year, and then generally has an opportunity
of selecting a bride — but before he can enter into any negotiations, he
must return home and inform the villagers of his intention and obtain
their sanction. The practice of referring a marriage to a tribal council
or a whole group is widespread. In New Zealand when a young man
married a girl of different tribe, failure to obtain the permission of both
might lead to war. Among the Hurons of North America ‘proposals
made to the girl’s mother were submitted by her to the women’s coun-
cil, whose decision was final’ ; while among the Pawnees a marriage was
discussed in a solemn assembly by all the relatives of both parties. In
Western Australia the consent of the whole tribe is necessary to the con-
clusion of a marriage, while in the Gilbert Islands all the males of the
clan must signify their consent, as also with the natives of San Cristobal
in the Solomons.

The presents which it is customary to exchange on the occasion of a
marriage, or the bride-price which is paid, are in many of the most
primitive groups supplied by all the relatives of the one group, and
distributed among all the members of the other group. Thus among the
North American Indians such presents are collected from the husband’s
kindred and are equally distributed among the members of the wife’s
family. Among the Araucanians all the relatives share the bride-price.
Among the Yakut there are two distinct bride-prices, one payable to the
bride’s parents and the other to the whole of her clan. In Samoa, the
immediate relatives of the bridegroom ‘had to go on begging expeditions
to all who were connected with them, and collect from them large
quantities of property’, and the family of the bride did the same. Similar
procedures are widespread in every part of Africa. The whole tribe
contributes towards the payment among the natives of Central Ceram
and the Toradjas of the Central Celebes.

Furthermore, a marriage is commonly celebrated not by the happy
couple but by the two families; indeed, the bride and bridegroom are
sometimes absent from the ceremony. They do not appear at the wed-
ding celebration among the Kirghis, the Yurok or the Buryat; while
among the Pathans the one person who is conspicuous by his absence
in the wedding procession is the bridegroom. The wedding banquet
might, at first sight, appear to be an institution calling for no explana-
tion beyond the natural one that it serves as an occasion for conviviality.
But closer examination suggests that there is more to it. Among many
peoples the eating of food together by the bride and bridegroom is the
essence of the marriage ceremony, and often it is the only occasion on
which they take a meal together. It seems possible that it is this rite

THE INSTITUTION OF MARRIAGE 117

which actually joins together the peoples in matrimony, 1 and this
conjecture is strengthened by the forms which the banquets sometimes
assumed. Thus among the Yakut the bride and bridegroom do not take
part in the banquet, but sit with their faces towards the wall. The
relatives exchange pieces of meat, symbolizing the union of the families
which henceforth are to be ‘flesh of one flesh, and bone of one bone’.
Among various tribes of British Central Africa, unless the parents of
both parties cat the food cooked by the bride’s parents, the marriage
is neither legal nor binding. Similarly among the Rajbansis of Purnea,
in Bengal, a man who is too poor to afford such a meal may marry a
woman with elaborate religious ceremonies, but until he has saved up
enough money to provide a wedding meal he is regarded as living in a
state of concubinage. The Chuhra of the Punjab do not admit the other
party to family privileges until after the ceremonial meal. The wedding
cake is, of course, an essential constituent of such banquets. Among the
Kolarian tribes portions of the cake are presented to the totems of the
respective parties. Certain clans which have trees for their totems offer
wedding cake to the sacred trees; others of the cobra-totem deposit
pieces of wedding cake in cobra holes, and so on. Certain Basors, who
express their religious emotion by violent dancing, partake of the wed-
ding cake while performing somersaults. It is of course obligatory for
all guests to eat some of the wedding cake, but among the Kols it
would be highly improper for any persons who are forbidden by law
to intermarry to do so, for they would thus be married and hence
become guilty of incest.

Elsewhere it is necessary to eat the wedding banquet hastily. Among
the Karubas of Southern India, should anyone choke during the pro-
ceedings or afterwards be seized with flatulence or indigestion, this is a
bad omen. Among the Yurok, a Samoyed tribe of Siberia, the wedding
guests must not only eat as hastily as possible but must depart immedi-
ately the last mouthful has been swallowed. It would therefore seem
that the banquet does not really have the character of a convivial
gathering, as we so generally assume, but rather echoes other ritual
meals of which we know, which are a means of effecting a communion
of flesh between the participants. In short, it would seem that such
banquets arc intended to unite, not only the bride and bridegroom but
their respective families.

Another common marriage rite is the marking of the bride and bride-
groom with blood, the contract being assimilated to the blood-bond
by which a person becomes adopted into the tribe. Now, in some
instances, not only are the bride and bridegroom thus marked but all
their relatives also, as in Cambodia and among the Papuan savages of

1 See also p. 262 .

Il8 THE MOTHERS

the western coast of New Guinea. In some parts of Polynesia, it is not
the parties concerned who mingle their blood but their respective
mothers.

Reasons for Marriage between Groups

Marriage, then, is almost universally regarded as a contract not between
a man and a woman but between the groups to which they belong. No
feature is more marked and more general, and it is even more pronounced
in the most primitive phases of society. The Blackfellows of Australia,
the wild Seri of Tiburon, are one with the Chinese and the French
nobility in viewing marriage as a matter primarily concerning two con-
tracting groups. Evidently, a contract between two groups must, in the
conditions of primitive society, have been a matter of far greater im-
portance than any possible individual relationship. These elaborate
negotiations can be better understood if we remember that, under the
rules against incest, a man must, if he marries, marry someone from
another group ; but in primitive societies there is very little intercourse
between groups, and this may be a very difficult task. The members of
one’s own group are, in primitive society, ‘our people’, all others are
enemies. Mr Mathew remarks: ‘When a Blackfellow crosses the boun-
dary lines of his own territories, he takes his life in his hands.’ Further-
more, clandestine visiting is even more difficult than open intercourse,
for almost all savages are restrained by superstitious fears from going
out of their camp after sunset. Unless there exists some friendly
understanding between the two groups, it is impossible to do so during
the daytime. For example, among the Bakyiga, a tribe inhabiting the
shores of Lake Edward, a man must, by the laws of exogamy, procure
a wife from one of the neighbouring clans, but, since all the clans are
in a perpetual state of deadly war, it is quite impossible for him to visit
another clan without a virtual certainty of being murdered. Negotia-
tions for marriages arc accordingly conducted by women only, and when
the bride comes to join her husband, she is brought by female relatives
only. There is, in fact, but one way in which such difficulties can be
overcome while observing the prohibition of incest; that is, by some
pact between the two groups, such as that which the sons of Hamor
proposed to the sons of Jacob: ‘Then will we give our daughters unto
you, and we will take your daughters unto us.’

This is what in fact takes place in primitive societies. The elaborate
tribal conferences and conciliatory exchanges of presents which have
been noted in Australia and Melanesia or Polynesia, do not, of course,
take place on the occasion of every marriage, but only when a marriage
takes place between members of two different tribes between whom no
such practice has become established. The negotiations which take

THE INSTITUTION OF MARRIAGE 119

place refer only incidentally to the individuals. Primarily they concern
the relationship between the two groups.

Thus it is that the original character and purpose of the juridic trans-
action of marriage is not to legalize the sexual relations of individuals
nor to safeguard the claims of individual possession; it is a contract about
the relationship of groups, rendered necessary by the rules against
incest.

The Marriage of Cross- cousins

Such an agreement between the two groups implies that every member
of each group shall marry into the other group. This relationship, if
continued long enough, results in an institution known as cross-cousin
marriage. Let us suppose, for the sake of example, that in each of two
primitive groups which we will call ‘Bears’ and ‘Wolves’, consisting of two
generations only , all the Bear men will be married to Wolf women, and
all the Bear women married to Wolf men. Then, if children are regarded
as belonging to their mother's clan, since the eldest male Wolf will be
married to a Bear woman, the father of every Bear will be a Wolf, and
similarly the father of every Wolf will be a Bear. The wives of the
younger Bear men will be Wolf women — that is, they will be the
daughters of their father’s sister, and they will also be the daughters of
their mother’s brother. In short, every Bear man and every Wolf man
will marry a daughter of his mother’s brother, and every Bear woman
and Wolf woman will marry a son of her mother’s brother. In other
words, all unions in the two groups will be between cross-cousins. (In
our own language we make no distinction between a father’s brother’s
child or a mother’s sister’s child, calling both first-cousins, but among
the vast majority of peoples a distinction is drawn between the two,
the first kind being known as cross-cousins, the latter as ortho-
cousins.)

Furthermore, among many peoples, while marriage between cross-
cousins is allowed, marriage between ortho-cousins is prohibited as
incestuous, those relatives being regarded as brothers and sisters. The
reason for this distinction is manifest: daughters of a father’s brother
or of a mother’s sister were members of the same group, or clan-sisters.
Nor is this all. Not only are marriages between cross-cousins encouraged,
but what is regarded as being a peculiarly desirable union is to marry
a girl who is at the same time the daughter of one’s mother’s brother and
of one’s father’s sister.

In Melanesia this system is the foundation of social organization.
Thus in New Britain a man is expected to marry the daughter of his
mother’s brother, but may on no account marry the daughter of his
mother’s sister. Among the majority of the more primitive populations

120

THE MOTHERS

of the Malay Archipelago, to marry a cross-cousin is regarded as a moral
obligation, and failure to do so is visited with a heavy fine. The fine is
similar to that imposed on anyone who has incestuous relations (as
they are considered) with his father’s brother’s or his mother’s sister’s
daughter. The same customs obtain in the various islands of the
Southern Moluccas and also in Ceram.

Cross-cousin marriage is a fundamental social law with a large
proportion of the aboriginal races of India. It appears to be widespread
among the primitive populations of Northern and Central Asia, al-
though owing to the ambiguity of our terms of kinship, the reports of
explorers do not always make it clear whether cross-cousins are meant.
The Aleuts, according to Father Veniaminoff, ‘marry by preference
the daughter of their uncle’. A similar preference is shown by the Ainu,
the Kamchadals, the Chukchi, the Koryaks and others. This view
obtains in several parts of Africa, for example among the Bakongo, the
Herero and others, while among the Bechuana the practice is so
common that it must almost be considered the general practice of the
tribe. These rules appear to have been common to all the Caribbean
races of Central America. ‘Marriage among the totem-societies of
Australia, America and India is both exogamous and endogamous; a
man is forbidden to marry either within his own clan or outside a
certain kinship group.’

As Dr Codrington very lucidly explains when speaking of the Melanes-
ians: ‘In the native view of mankind . . . nothing seems more funda-
mental than the division of the people into two or more classes which are
exogamous, and in which descent is counted through the mother. This
seems to stand foremost as the native looks out upon his fellow men;
the knowledge of it forms probably the first social conception which
shapes itself in the mind of the young Melanesian of either sex, and it
is not too much to say that this division is the foundation on which the
fabric of native society is built up.’ Sometimes a clan is divided into
two groups, the members of any one group being prohibited under
pain of death from intermarrying and being obliged to draw their
sexual partners from the other group. Or instead of two exogamous
groups, there may be four or even eight. Totemic considerations may
also apply, so that the restrictions become very complicated. Since
marriage between members of the same village is regarded as incest, it
often happens that, as among the Zayeins, the unmarried males and
females dwell separately in two large houses at the opposite ends of the
village, and so afraid are they of incurring suspicion that any amatory
relations exist that they are careful to lower their eyes discreetly should
they happen to meet. If such a crime comes to light the culprits are
made to dig their own graves, then they are pushed in and buried alive.

THE INSTITUTION OF MARRIAGE 121

It can cause no surprise to learn that the Zayeins are steadily diminishing
in numbers.

The custom of cross-cousin marriage is now well enough understood
for it to be unnecessary to adduce further evidence here. As Dr Rivers
remarks: ‘It would seem impossible to find any direct psychological
explanation in motive of any kind, whether religious, ethical or magical.
They [the customs] seem to be meaningless except as a vestige of an old
social order, while when considered from that point of view they be-
come at once intelligible and natural.’ It has been suggested that cross-
cousin marriage may be explained by the economic advantage of keeping
family property undivided. This cannot apply to the savages of Australia,
Melanesia or the Caribbeans, who transmit no family property. More-
over, it is patriarchal societies which care most about property, and yet
few such societies practice ortho-cousin marriage; it is virtually con-
fined to the Arabs and some other Muhammadan peoples of Egypt,
the Sudan, Morocco and India, and in each of these cases it is said to
be adopted precisely with the object of keeping property in the family.
The ancient Greek ‘laws of Gortyna’ make the matter very clear: ‘An
heiress shall be married to the brother of her father, to the eldest of
such brothers as there may be. If there be several heiresses and several
brothers let each of the younger ones be married to one of the brothers
in the order of their ages. If there be no brothers of her father . . . then
let the heiress marry the son of the eldest brother.’ Inheritance through
the daughters is here adapted to patriarchal aims, not by cross-cousin
marriage but by ortho-cousin marriage. It is, in fact, unlikely that a
custom devised to keep the property of a matriarchal family undivided
should be preserved under a patriarchal system — the chief purpose of
which is to abolish such a mode of succession.

Bakongo society offers a striking illustration of the essentially matri-
archal character of clan organization and the patriarchal character of
the family group. The exogamous clans are reckoned to have sprung
from the offspring of a common ancestress, and descent is reckoned
in the uterine line, yet the father is the absolute head of the family and
the property is transmitted in the male line. Nevertheless, the Bakongo
strictly observe cross-cousin marriage. A similar paradox is found
among the Rabhas of Assam. Often economic advantage is set aside, as
among the Bataks of Sumatra, a severely patriarchal people who never-
theless set advantageous alliances aside in order to make cross-cousin
marriages.

Westermarck makes it a principle that customs should be explained
by existing conditions and not as survivals of hypothetical past condi-
tions. Quite to the contrary, a sounder rule would be: whenever a usage
of world-wide distribution is found to be observed by races of widely

122

THE MOTHERS

different cultural levels, the explanation of its origin is to be found in
interpretations which apply to the least-advanced of these peoples.
Westermarck’s principle leads to error since plausible current reasons
can always be found — for instance, the explanation of the rule of ex-
ogamy as a provision against the injurious effects of inbreeding. On
this principle, we should have to interpret the observance of sabbath
days as due to the advisability of regular rest intervals, whereas we
know the explanation to be quite different . 1 Similarly, we might think
mourning customs prompted by respect for the dead when in fact they
are prompted by dread of the spirit of the departed. In modern society
no one supposes that such customs as raising one’s hat to a magpie are
to be explained by reference to existing conditions.

The Change from Clan-kinship to Family kinship

Once the primitive maternal clan organization has broken down and
kinship comes to be viewed in relation with another form of group, i.e.
the family, both ortho- and cross-cousins are regarded as equally close
relatives, and marriage of either may be seen as incestuous. Thus many
societies condemn marriage of first cousins of either sort and even of
second, third, fourth and more remote ones, so that the very relation-
ship which is prescribed under one system is condemned under the
other. The transition is clearly seen in India where among some tribes
and castes the one form of marriage is an obligation, in other commend-
able, in others permissible, while yet others condemn it. Sir Denzil
Ibbetson writes of the Punjab: ‘The people are beginning to add the
mother’s mother’s clan to those in which a man is forbidden to marry
or even to substitute it for the father’s mother’s clan, and this is appar-
ently the last stage in the change from relationship through women to
relationship through men.’

The clan-group organization tends to break down in time into family-
group organization. Since exogamous clans or classes are general
throughout Australia, Melanesia and primitive Indonesia, there can be
little doubt that they were once general in Polynesia, but they have
vanished. Again, the Yakut were once so strictly exogamous that blind-
ness was thought to be visited upon the trespasser, but today a man
marries any woman outside his own family. Similarly with the Iroquois.
Moreover, even where, as in Fiji, the clan organization has fallen into
complete decay, cross-cousin marriage may persist.

Primitive Kinship and Group-motherhood

We must dismiss the idea that while the family is a group depending
upon reproductive and economic relations, the clan is a group resting
1 See below, p. 250 ff.

THE INSTITUTION OF MARRIAGE 123

upon some other principle of a social or political character. Primitive
humanity is innocent of political organization, and organization through
the authority of a chief is a late phenomenon. ‘Cohesion in a community’,
remarks Mr Mathew, in speaking of the Australians, ‘depends entirely
upon consanguinity, and derives no strength at all from governmental
authority.’ Primitive kinship terms support this idea; all male members
of the clan of about the same age, including those we should call cousins,
are reckoned as brothers, and all females as sisters; the older male
members are called by a term which includes both ‘father’ and ‘uncle’;
the older female members by a term including both ‘mother’ and ‘aunt’.
The younger members of the corresponding marriage class are called
‘wives’ and ‘husbands’, i.e. a wife’s sisters or a husband’s brothers. This
is the so-called classificatory system, as against the system which obtains
in our own society which is known as the descriptive system. This label
assumes that our own system is somehow natural. We tend to think that
when the term ‘mother’ is applied to all women of the same generation
as the actual mother that this is a convention but, as Mr Thomas ex-
plains, when we find that the term which is translated ‘son’ is applied
not only by the biological mother but by all the other women of the
group to a particular boy, we ‘may discard the hypothesis that “wife”
means wife, “husband” husband, and conclude that if there was a period
of group marriage there was also a period of group motherhood. This
interesting fact may be commended to the attention of zoologists.’

But zoologists would tell him that there is an absence of any relation-
ship between the operation of the maternal instinct and actual con-
sanguinity among animals — as we know from the case of the cuckoo and
many others. Thus filial and fraternal instincts operate in accordance
with the classificatory and not with the descriptive system of relationship,
and it is the former which is, in the biological sense, natural. The idea
that a system of relationship must rest upon the biological fact of genera-
tion arises from the fallacy of ascribing to primitive ideas an intellectual
character which is foreign to them. This fallacy is embodied in the very
word ‘system’ — it is assumed that the reckoning of kinship is based on a
deliberately devised and intellectually sound scheme. That this is
equally true of humans is shown, for instance, by the reply of an Iro-
quois to Father Lejeune when he asked him how he could be so fond
of children of whose parentage he could not be sure. The Indian looked
at him contemptuously and replied: ‘You Frenchmen love only your
own children, we love all the children of the tribe.’ Wives who attend
to their own children and to those of other women without distinction
are widely found among primitive races, and it is an ‘almost universal
custom that a child should be suckled not only by its mother but also by
its grandmother, and by other near female relatives’. In many cases foster-

THE MOTHERS

124

relationship systems are maintained and children are interchanged. In
the Hindu Kush these foster relationships are regarded as so close that
marriage between foster relations would be looked upon as incestuous.
If a man is suspected of adultery, any further misconduct is insured
against by obliging him to put his lips against the woman’s breast — she is
henceforth regarded as his foster-mother, and no other relationship but
that of mother and son can exist between them. These fostering usages
are common in India, are prevalent in primitive Indonesia, especially
among the Dayaks, and are found in Polynesia and Melanesia. In the
Gilbert Islands the persons who pass for a man’s or woman’s parents
are never the real parents, for every child is adopted at birth by foster-
parents, as soon as suckling is terminated. The children are so com-
pletely assimilated as to be unconscious of their relationship to their
real family. Similar customs are found in West Africa, as well as among
such tribes as the Aleuts and the Cheyennes.

Yet Dr Westcrmarck exclaims: ‘Where in the world has a society
been found in which it is the custom for infants to be taken away from
their mothers when weaned, or for mothers to desert their infants ?’ His
inquiry is all the more remarkable inasmuch as the usage is found
among some of the peoples of Morocco, a country about which he
claims a special knowledge.

To summarize, the aspect of the clan which is foremost in the primi-
tive mind is the sexual aspect; all primitive marriage regulations are
dominated by the desire to avoid incest. To the Australian, the Melanes-
ian and the North American Indian the primary purpose of clan divi-
sions is to indicate who may marry whom.

Primitive Marriage Relationship

These inter-marriage rules refer exclusively to the relations between
groups, and are not designed to safeguard claims to individual sexual
possession — which we regard as the main purpose of marriage laws and
morality. The relationship of individual marriage is denoted by no
specific term in primitive nomenclature. A wife’s sisters and a husband’s
brothers are not merely called wives and husbands, but act and are
treated as such, as we shall see in the following chapters. In most cases
such usages are associated with some form of individual marriage, i.e.
the continuous cohabitation of one man with one or more women. But
this association does not necessarily exclude collective sex-relations.
The sexual and economic aspects of association are, in fact, quite inde-
pendent. Individual marriage rests upon the economic relation, for a
man may contribute the products of his hunting or his services as a
warrior to the whole group, but the economic value of a woman lies in
her personal service. To obtain this service is the chief motive of

THE INSTITUTION OF MARRIAGE 125

individual marriage in most primitive communities. Jealousy among
most primitive races refers not to exclusive sexual possession but to
economic loss, and claims to exclusive sexual possession have developed
out of this.

In the societies which have preserved their primitive organization,
freedom of access between any male of the one group and any female of
the other is the rule rather than the exception. Where individual sexual
rights exist, penalties for infringing them, i.e. for adultery, are usually
much less severe when the infringer belongs to the same group as the
husband than when he does not. Indeed, in the former case there may
be no penalty : a temporary loan or exchange of wives between members
of the same marriage group is quite common.

Nevertheless, complete promiscuity is nowhere found as an accepted
social pattern. Though sexual relations are, in primitive societies, often
loose and casual they are by no means unregulated, for the incest pro-
hibitions both exclude certain relationships within the group and
necessitate an agreement with some other group. To assess this early
form of the marriage institution by current standards of sexual morals
and to regard it as approaching promiscuity — as many missionaries
have done — is unscientific and inaccurate. People always feel that sys-
tems different from their own are barbarous. Thus the polygamous
Singhalese spoke of the wild Veddahs, who had only one wife, as being
‘like monkeys’, while in South Africa coastal natives who have given up
such practices speak of the inland tribes as ‘no better than dogs’, in this
respect. The Jews looked upon the marriages of Christians as wholly
invalid.

Far from being unregulated, Australian and Melanesian sexual rela-
tions are subject to much more elaborate regulations than our own, and
these may make the choice of a wife much more restricted than in
modern society, or may even make it impossible. These rules may
become, as among the Zayeins of Burma, an intolerable tyranny. As a
matter of fact, if we seek for the nearest approach to promiscuity we find
it in societies which arc formally ‘monogamous’, such as some forest
tribes of Brazil and Malaya.

To recapitulate: marriage originated in a contract between two kin-
ship groups, and only later became a contract between individuals,
when clan kinship broke down into family kinship. In the next two
chapters let us pursue this novel idea further, examining every quarter
of the world for evidence of group marriage.

CHAPTER 10

Group Marriage (i)

Sororal Polygyny

I T is a widespread principle in uncultured societies that when a man
marries a woman he thereby acquires marital rights over all her sisters.
Thus, among the Kurnai of South-east Australia, when a man elopes
with a woman from another tribe, her parents, after their anger has
blown over and the matter has been amicably settled, hand over her
sister also. Among the tribes of Gippsland the men cannot be made to
understand the distinction between a wife and a sister-in-law ; the latter,
they insist, are just as much their wives as the former. So also in the
western islands of the Torres Straits before conversion of the natives to
Christianity, a man’s wives were all sisters or cousins, and even today
there is little doubt that husbands normally have marital relations with
the wives’ sisters. This principle, like the rule of cross-cousin marriage,
is a translation into terms of family relationships of the sexual claim of
a man to all the women in the group with which his own group has
entered into a marriage agreement.

This arrangement, which is called sororal polygyny, was almost uni-
versal among the North American tribes, and appears to have been
equally general in Central and South America, where it is found from
the Canebo of the Upper Amazon to the inhabitants of Tierra del
Fuego. It is very common in Africa (e.g. among the Zulus and Kaffirs),
especially among the primitive races, being a matter of course among
the Upper Congo tribes.

It is also usual among the primitive races of Siberia, and is old-estab-
lished among the Mongols and Chinese. Jinghis Khan married two
sisters, and the practice was taken for granted among his warriors and
khans. We read of the Emperor Yao bestowing both his daughters on the
Chinese prince Sheunn. Among the ancient Japanese, ‘to wed two or
more sisters at the same time was a recognized practice*. We also find it
in Tibet, Cambodia, Siam, among the Malays of the Patani States, and
among the tribes of Upper Burma and Manipur. It was in vogue among
the ancient Indo-Aryans, and is common in the Punjab, as well as
among various native tribes of Central India, and is prevalent in Mysore
and Southern India. We know from the account of the marriage of
Jacob that it was a recognized usage among the ancient Jews.

GROUP MARRIAGE (i) I2J

In the Pacific area, similar evidence comes from the Philippine
Islands, the Marshalls, the Gilberts, the Mortlocks and elsewhere.

In practice, sororal polygyny is limited by the factor which limits
every kind of polygamy — the financial strain of marrying a whole family.
This difficulty is often relieved by the fact that, since in primitive
society girls usually marry at puberty, the younger sisters are not yet
marriageable when their elder sister marries. By the time they are
eligible the man’s circumstances may have improved, enabling him to
maintain a larger family. Hence the rule, observed in most parts of the
world, that the younger sister shall not marry before her older sister — a
rule on which Laban insisted when he gave his daughter to Jacob, and
one which has left traces even in England and Scotland.

If a husband does not wish to exercise his claim on his wife’s sisters,
he may allow them to marry other men and, in some cases, may receive
the bride-price instead of the girl’s relatives. An additional sister is
given as a matter of course if the first sister proves barren. Among the
Tartars, if the wife dies before the payment of the bride-price is com-
pleted, the sum already paid goes towards the acquisition of her sister,
but if there is no sister to take her place, the deposit is lost. In other
instances, the bridegroom may demand a refund of the bride-price.

Among the Flatheads and other Oregon tribes, when a man’s wife
dies, if her sister is already married to another man she is obliged to
leave him and marry the widower. Among the Wabemba of the Congo
in a similar situation, the husband of one of the deceased wife’s sisters
must allow his wife to cohabit for one or mo nights with the widower.
Unless this is done, the latter cannot remarry. If the sister is still an
infant, she is nevertheless handed over to the widower, but a slave-girl
is sent with her to act as a substitute until she is of marriageable age.
Similarly among the Baholoholo, who observe the same rule, if the
surviving sister is a mere infant the widower goes through the form of
imitating the sexual act with her. (We shall see that similar ritual sur-
vivals are associated with the corresponding custom of fraternal succes-
sion.)

Marriage with the deceased wife’s sister is sometimes regarded as a
moral obligation rather than as a claim or privilege. Among the Shuswap
of British Columbia, the widower was actually kept prisoner by his
deceased wife’s family, and was released only on condition that he
married the sister. The notion that it is reprehensible to marry one’s
deceased wife’s sister is an anthropological curiosity which appears to
be found only among the natives of New Britain, some Chinese tribes
and some natives of Ashanti. Nevertheless, the people who observe this
rule but who have given up sororal polygyny do not admit that they
practise the former custom because they once practised the latter, but

128

THE MOTHERS

give some specious explanation. In the same way, certain anthropo-
logical writers to whom the application of the theory of evolution to the
human race is repugnant have no hesitation in declaring that they can-
not ‘find any reason for the assumption that the custom of marrying a
deceased wife’s sister is derived from the custom of marrying her other
sisters in her lifetime ’. 1

The favourite explanation given by travellers and missionaries who
last observed the custom is that it is desirable in a polygynous family
that the wives should be sisters because sisters are more likely to live
together in harmony. In fact, where polygyny obtains, the women are
the most persistent advocates of the practice, and there is nothing to
indicate that the wives in polygynous families are more prone to quarrel
among themselves than other persons who live together. Indeed,
according to La Potherie, among the North American Indians sisters are
particularly quarrelsome and sometimes attack one another with knives;
while the Ostyak, who have immemorially practised sororal polygyny,
state that the arrangement is unsatisfactory since sisters are particularly
liable to disagree in such marriages.

Actually it may be doubted whether Melanesian savages are much
concerned about the amicable nature of the relationships between their
wives, about respect for their deceased wives, or the proper qualifications
in the nurses of their children — to name some supposed reasons for the
practice. In any case, none of these alleged benefits account for a man
having to marry his wife’s sisters against his will, or for his collecting
the bride-price when they marry other men, or for his having to wait
with a slave-girl as a substitute when those sisters are still infants-in-arms.

Like cross-cousin marriage, sororal polygyny is a translation in terms
of family relationship of the wider conceptions of clan relationship.
From the point of view of the clan-group, the term ‘wife’ includes all
women of the corresponding marriage-group, and all those women are
sisters. Dr Codrington, speaking of the Melanesians, declares that, ‘to
a Melanesian man, all women, of his own generation at least, are either
sisters or wives’ [actual or potential], and that ‘all women who may
become wives by marriage and are not yet appropriated, are to a certain
extent looked upon by those who may be their husbands as open to
more or less legitimate intercourse. In fact, appropriation of particular
women to their husbands, though established by every sanction of native
custom, has by no means so strong a hold in native society, nor in all
probability so deep a foundation in the history of the people, as the
severance of either sex by divisions which most strictly limit the inter-
course of men and women to those of the section or sections to which
they do not themselves belong . 5

1 Westermarck: History of Human Marriage , III, 263.

GROUP MARRIAGE (i) 129

The complement of the rule of sororal polygyny is that of fraternal
polyandry; i.e. that when a woman contracts a marriage with a member
of another family she marries all the marriageable males of that family.
The simultaneous observance of both these rules constitutes a marriage
between the two groups, irrespective of the relations between the
members composing them. However, the one-sided observance of
sororal polygyny, and perhaps also of fraternal polyandry, are much
more common today than the combination of the two practices as
complete group-marriage. The reason for this is plain. The combination
of the two practices is an unstable arrangement, and can only operate in
a modified form where sexual relations do not entail permanent co-
habitation. As soon as marriage involves economic interdependence
the arrangement becomes impracticable, for economic association can
only take place when the labour of the woman or women is in some
degree specially allotted to the husband, and it necessarily breaks up
into one or another of its aspects — that is, into sororal polygyny or
fraternal polyandry. But since the development of individualism and
individual property has taken place in the hands of the men, it is natural
that when it breaks up it does so in the form of sororal polygyny. In
point of fact, fraternal polyandry, though geographically widespread,
is considerably less common than sororal polygyny, being rarer than
is generally supposed. Many of the customs usually described as
fraternal polyandry turn out, on inquiry, to be associated with sororal
polygyny and thus to represent, in reality, group-marriage. We shall
now survey the two subjects together.

Northern Asia

The confusion to be found in many early anthropological reports is
well illustrated by the Gilyak. As the careful investigation by Dr
Sternberg has shown, they are strictly organized into exogamic classes,
and every member of each marries into his allotted class on the basis
of cross-cousin marriage. Individual ‘marriage’ takes place in the sense
that a woman becomes the particular economic associate of a man but
her economic husband possesses no sexual rights over her. 1 Earlier
reports misunderstood the position. An old Japanese traveller says
that Gilyak women had several husbands, while another traveller reports
that their sexual relations are indiscriminate.

Of the Chukchi of extreme North-east Asia a Russian traveller
mentions that ‘among other customs they have the usage of contracting
so-called “exchange-marriages”. Two or more men enter into an agree-

1 Among the Western Gilyak all tribal brothers have marital rights over the
wives of each other indifferently, but among the Eastern Gilyak the older
brothers have no right to the wives of the younger brothers.

t.m. — 5

THE MOTHERS

130

ment whereby they have mutual rights to each other’s wives. This right
is exercised whenever the contracting parties come together, as for
instance on the occasion of a visit.’ However, the elaborate monograph
of Mr Bogoras makes clear that the Chukchi are commonly betrothed in
their infancy to their first-cousins, and that they observe sororal poly-
gyny. In this instance we have clear testimony that licentiousness has
nothing to do with the institution. The Chukchi are indeed described
as a sensual race, but their group-marriage organization is not exploited
for licentious purposes. They arc careful not to form such an alliance
with persons dwelling in the same village, and in general avoid exercis-
ing the rights it confers on them. Sometimes cousins exchange wives
for several months, for years or permanently. So seriously is the
arrangement regarded that children in the same marriage-group are
regarded as brothers and sisters, and are not allowed to marry among
themselves. It appears that, although the group-marriage of the Chukchis
is to a certain extent artificial, since it depends upon an individual
compact, it nevertheless is a derivative of established marriage-rights
between two marrying classes or groups, modified by the necessities
imposed by their isolated and scattered way of life.

This view is confirmed by the position among the Aleuts, where
thorough investigation has been made by Mr Jochelson, who says:
‘To participate in group-marriage is the duty of cousins.’ Here, again,
earlier reports described the system as consisting of loose polyandrous
unions.

Sexual Hospitality

Let us pause for a moment to consider how it is that participation in
group-marriage, which we tend to regard as a form of licence, should
be regarded not only as a right and a privilege but as a moral obligation.
Community of wives was originally regarded as essential to the relation-
ship of tribal brotherhood. To primitive man all men are either tribal
brothers or strangers, and the latter term is equivalent to ‘enemy’. If
a man, not being by birth a tribal brother, is admitted into a community
and is regarded with good-will — in short, if he is not an enemy — he
must needs be a tribal brother. Hence the sacredness of hospitality in
all primitive sentiment. If a man has touched the tent-rope of an Arab’s
tent, his life must be defended against all enemies. The hospitality of
savages knows no bounds. If they are on the verge of starvation, they
will give what little they have to the stranger who has been admitted to
their midst. The first thought of the savage, when the stranger to whom
he feels himself attracted is in his company, is to take the necessary
steps to make him a tribal brother. In Australia, if the member of a
strange tribe refuses to drink the blood of his hosts, it is forcibly

GROUP MARRIAGE (i) 131

poured down his throat. Among the Koryak the guest is obliged to
undergo a somewhat strange rite of brotherhood with his host’s wife
before he can avail himself of her hospitality. ‘Elle lache son urine en
presence de l’etranger et lui en offre une jatte pour s’en rincer la bouche.’
It follows that the host’s lending his guest his wife is not misguided
benevolence but a necessary pledge of friendship. For the guest to
refuse this so-called ‘hospitality prostitution’ is the equivalent to
repudiating the assumed brotherhood and is thus tantamount to a
declaration of war. The sedentary Koryak ‘look upon it as the truest
mark of friendship when they entertain a friend to put him to bed with
their wife or daughter; and a refusal of this civility they consider as the
greatest affront, and are capable of even murdering a man for such
contempt. That happened to several Cossacks before they were acquain-
ted with the customs of the people.’ The same is reported of the Chuk-
chi. In Madagascar a missionary closely escaped being murdered
because he refused the proffered hospitality. The custom is very general
in all primitive societies . 1

Thus it is that the existence of clan-brotherhood is, or was, formerly
considered to imply sexual communism. Naturally the practice of sexual
hospitality has tended to become modified in the same way that sexual
communism has become modified by the development of individual
marriage. All manner of transitional forms are found. Thus, the Mis-
souri Indians were so averse from intercourse with members of another
tribe that they never offered their wives or daughters to strangers.
Nevertheless, they regarded themselves as under the obligation of
offering sexual hospitality to a guest, and accordingly provided him with
a captive from another tribe. It may safely be concluded that this was a
compromise between their endogamic tribal principles and their equally
strong conviction that a clan-brother was entitled to access to his
host’s women. Again, among the Krumir Berbers a stranger visiting
the tribe is lodged by one of the tribesmen, and is invited to spend the
night in the company of his host’s wife. The host mounts guard outside
the tent, and should he hear any suspicious movement would have no
hesitation in instantly shooting his guest. So-called ‘hospitality prosti-
tution’ has here dwindled to an empty ceremonial. Similarly, since the
Arabs attach great importance to hospitality and clan-brotherhood, we
may infer that at some previous period sexual communism among clan-
brothers was customary. And in fact the learned Arab jurist, Ata
ibn-Abi Rabah, states that the custom of offering one’s wife to a guest
was of old a universally-sanctioned custom of the Arabs. In some
Arab tribes this has survived down to the present, or quite recent, times.
Thus the clan organization of a people whose notions of individual
1 A bibliography of 116 titles is appended.

THE MOTHERS

132

marriage are today even severer than our own nevertheless entailed the
same conceptions as among the most primitive savages.

Collective Sexual Relations in America

There can be little doubt that the practice of exchanging wives tempor-
arily, universal with all sections of the Eskimo race, is a survival of
tribal sexual communism, which has become disintegrated by the dis-
persion of these small communities. In Repulse Bay, ‘it is the usual
thing among friends to exchange wives for a week or two about every
two months’; and Dr Murdoch was informed that ‘at certain times
there is a general exchange of wives throughout the village, each woman
passing from man to man until she has been through the hands of all’.
In Northern Greenland, as Dr Bessels delicately puts it, ‘somewhat
communistic tendencies seriously interfere with the sanctity of mar-
riage’. The most important race of the extreme north-western region
of America are the Tlinkit. They are divided into a number of totemic
clans grouped into two large exogamic marriage classes. Polygamy is
very general, a man of distinction having as many as forty wives; the
Tlinkit are also polyandrous. Here again we see the deceptive manner
in which such an organization is apt to be reported. Some writers report
that their customs allow ‘great looseness in sexual relations’. Count
Langsdorf, on the contrary, says that their decent and orderly behaviour
and the modesty of their women stand in striking contrast with the
manners of neighbouring races. Adultery is severely punished, but it is
only a serious offence if the seducer belongs to a clan other than the
husband’s. The rules governing their sexual organization seem fairly
clear, but the Russian bishop, Father Veniaminoff, feels it necessary to
go to Sicily for a parallel; many ethnological writers have taken the
hint and term these arrangements ‘cicisbeism’. It does not appear that
the customs of the Tlinkit bear any resemblance to the practices of
eighteenth-century Italian society, of which the relations between
Nelson and Sir William and Lady Hamilton are a famous instance;
they resemble far more closely group-marriage. It should be noted that
their marriages were matrilocal.

Making due allowance for the difficulty of observation, it would
appear that, from Manchuria on the Asiatic side to British Columbia
on the American, the principles governing sexual relations are substan-
tially identical among all tribes, and with the large majority, reciprocal
sexual rights between all the males and all the females, through inter-
marrying groups, are recognized, or were so until quite recently. The
general prevalence of those customs among the peoples of the ruder
north and west suggest that they may also have obtained among the
North American Indians, a presumption confirmed by several charac-

GROUP MARRIAGE (i) I33

teristics, such as the existence of the classificatory system of clan
relationship, and the practice of sororal polygyny which, as we have
seen, is universal among the North American Indians. In conjunction
with it, levirate marriage is a rite and an obligation . 1 Father Charlevoix
adds that, among the Senecas, who practise sororal polygyny, there is ‘a
far greater disorder, namely a plurality of husbands’. Thus there is
ample evidence to confirm the presumption that the marriage customs
of North American Indians were the result of decay of clan organization
under the influence of individual economic marriage.

Similar principles appear to have been widely current in South
America, and are reported of the natives of New Granada, the tribes
of Brazil, the Zaparos of Ecuador and many others. In Guiana, at the
present day, polyandry is common, and when a missionary endeavoured
to persuade an Indian to give up polygyny by asking him what he
would think of a woman having several husbands, he replied that both
customs were equally honoured in his tribe.

Collective Marriage in Tibet

Although Tibetan polyandry is constantly referred to and was used by
McLellan as the type of fraternal polyandry, the polygyny which goes
with Tibetan polyandry is overlooked by most writers. However, Mr
Savage Landor, who had exceptional opportunities of obtaining first-
hand information, gives us the most definite account. ‘A Tibetan girl
on marrying does not enter into a nuptial tie with an individual, but
with all his family. ... If an eldest son marries an eldest sister all the
sisters of the bride become his wives. ... At the same time, when the
bridegroom has brothers they are all regarded as their brother’s wife’s
husbands, and they one and all cohabit with her as well as with her
sisters if she has any.’ Tibetan marriage is, therefore, not simple poly-
andry but a complete group-marriage.

On the current assumption that polygyny and polyandry are different
and opposite forms of marriage, the mistake has occurred of describing
Tibetan group-marriage sometimes as polyandry and sometimes as
polygyny. But all reports agree to the existence of both these customs
in Tibetan-speaking countries . 2 Mr Landor was surprised by the rule
of seniority, by which, if an elder son marries the second sister, then

1 See pp. 153-6, for an account of the ‘lcvirate’.

2 For reasons presently to be noted, limitation of the number of wives appears
on the whole to be more common in the higher uplands, while plurality of wives
is more prevalent in the lower valleys ; and in one and the same valley polyandry
is stated to obtain in the upper portion and polygyny in the lower. It seems clear,
however, that the polygyny of the lower valley is, in reality, part of the same
organization as the polyandry of the upper.

THE MOTHERS

134

only the sisters from the second down become his property; but this is,
in fact, a characteristic feature of group marriage. It is understandable
when we realize that when an elder brother marries he does so as the
representative of a whole group of brothers. The younger brothers can
make no valid contract except through the medium of their elder. The
elder brother inherits the father’s property, and is under the obligation
of supporting his brothers. Since, on the marriage of an elder brother
his younger brothers have rights over his wives, it follows that any
marriage contracted by a younger brother independently would be
supererogatory. It would, in the Tibetan view, be bigamous.

Similarly among the Brahmans of Travancore and Malabar, when an
elder brother marries all the younger brothers arc debarred from con-
tracting any regular marriage within their own caste. The same prin-
ciple is observed in that residue of fraternal polyandry known as the
levirate, the right to marry a deceased brother’s wife. But while a younger
brother has this right as regards the widow of his elder brother, there is
no converse right of an elder brother to the widow of his younger
brother. The same is true of other tribes such as the Zulus and the
Thonga of Mozambique. On the same principle it is considered wrong
for a younger brother to marry before his elder brother. In China any
infraction of this rule is atoned for by carrying in the bridal procession a
pair of trousers, representing the elder brother, and laying them on the
chair of the bride.

Since there is every reason to believe that group-marriage derives
from an organization in which the marrying groups were originally clans,
or in later stages marrying classes, it is interesting that traces of these
classes have not completely disappeared in Tibet. Father Desideri, a
Jesuit, who visited the country at the beginning of the eighteenth cen-
tury, reports that there are two different kinds of kinship in Tibet,
kinsmen £ of the same bone’, and kinsmen ‘of the same flesh’. The first
kind comprising all individuals issued from the same stem, whether
begotten in the direct line or belonging to side branches however
remote. Those, on the other hand, designated as kindred of the same
flesh are matrimonial allies. The first type of kinship is regarded as
absolutely inviolable. This implies that the proper wife of a young
Tibetan is his cousin, which is confirmed by the fact that no marriage
can be concluded without the authority of the bride’s maternal uncle —
who would normally be the father of the bridegroom. Thus Tibetan
society, although essentially patriarchal at present, is not far removed
from a matriarchal phase. Formerly it was under a complete system of
gynaecocracy. The high position occupied by Tibetan women, in marked
contrast to the women in India or China, has already been noted. By a
strange paradox, despite the importance of the patriarchal household

GROUP MARRIAGE (i) 135

today, the patriarchal family has no name, and children are named after
their mother, not their father, whose name indeed must not even be
mentioned. In some districts polyandrous Tibetan marriages are also
matrilocal. Marriage is not regarded by the Tibetans as a sexual but as
a social and economic relation, the two aspects being kept distinct.
Marriage is regarded rather as a sacrifice in the interests of the family
group. Perhaps this is what Captain Turner meant when he wrote that
‘marriage amongst them seems to be considered rather as an odium and
a heavy burden, the weight and obloquy of which a whole family are
disposed to lessen by sharing it among them.’ (The question about which
Warren Hastings was curious to obtain information — namely, to which
of the fathers the children were supposed to belong — is entirely irrele-
vant.) It has several times been stated by older writers that polyandry is
peculiar to the poorer classes and is not practised among the well-to-do.
This is quite untrue. 1

The polyandrous organization of Tibetan marriage is not imposed
upon the women, who are staunch supporters of the system.

It is worth noting that the degree of sexual promiscuity is small,
except where group-marriage organization has been undermined by
individual marriage. Frequently, the several husbands arc absent from
home for long periods, and in practice a woman seldom cohabits with
more than one man at a time. Group-marriage is indissoluble, and the
easy dissolution of marriage owes its introduction to the infiltration of
marriage customs of the plains. Dr Westermarck takes occasion to
remark that — ‘in decay, polyandry is modified in directions tending
towards monogamy’. But, in point of fact, it is not towards monogamy
but towards the highest degree of polygamy that group-marriages
actually lead. The Marquis Cortanzc, impressed by the high moral
character of the Tibetans, remarked: ‘There does not exist in the whole
province of Ladakh ... a single Tibetan woman who is a prostitute;’
whereas the lowlands, where Tibetan customs have decayed, swarm
with them.

In Tibet there is no tradition of any origin of group-marriage, which
is regarded as ‘indefinitely old’, and these institutions have no connec-
tion with the Lamaistic Buddhism which was introduced at a com-
paratively late date. Mr Wilson reports that polyandry is almost uni-
versal all through Chinese Tibet and in almost all Tibetan-speaking

3 No importance can be attached to the various theories put forward ‘to explain
the origin 5 of so-called Tibetan polyandry. For instance, it has been said that it
is due to the disproportion between the sexes, though there is no evidence of
such disproportion. Mr Rockhill and other writers are of the opinion that ‘the
numbers of women and men are probably equal’, while in Lahul, where poly-
andry is extensive, the censuses of 1881 and 1891 show that women actually out-
number the men.

THE MOTHERS

136

provinces. It is probably the common marriage custom of at least thirty
million people in this area.

Polyandry is found throughout the Himalayan region in various stages
of decay and obsolescence. It is general in Saraj, the Simla hills and the
valleys of the western Himalayas. The gradual decay of polyandrous
institutions in the lower Himalayan valleys appears to be due, not to
changes in ethical conceptions but to the disintegration of the old social
organization. It is bound up with the communal undivided economic
household, in that all brothers remain in one family and share alike.
With the rise of individual economic interests, the polyandrous marriage
is necessarily broken up also. The highland regions of the Himalayas are
but a cultural island preserving social customs which once were far more
extensively distributed. Biddulph considers that polyandrous marriage
was once general in Hindu-Kush and Chitral, a view substantiated by the
eleventh-century Arab traveller, Al-Biruni. In Turkestan, according to
a thirteenth-century Chinese geographer, polyandry was obligatory. It
is curious that in the exogamic clans of China the same terms are used
as in Tibet to indicate distinction between kinsmen of the same flesh
and of the same bone; and sororal polygyny was once as customary in
China as with the ruder Mongol tribes of Central Asia. Today, to marry
the widow of one’s brother is so criminal an act that the guilty man is
condemned to be strangled. Nevertheless, it was at one time an estab-
lished usage among the Chinese.

Polyandry in North India

Polyandrous institutions were common in many parts of India, and
attracted the attention of the English when they colonized the mouth of
the Ganges. These people ‘stretch the Levitical law so that a brother
not only raises up seed to another after his decease, but even during his
absence on service, so that no married woman lies fallow.’ The writer
was probably referring to the Khonds — the most important of the non-
Aryan tribes in India. Similarly amongst the Santals, also representatives
of the aboriginal race, man has right of access to all the younger sisters
of his wife, and so have his younger brothers. In other words, there is
complete group-marriage. In addition, a Santal uncle ‘is permitted a
good deal of freedom of intercourse with his wife’s nieces’, a feature
which derives from the age-grade organization.

Attempts have been made to ‘explain’ polyandry by suggesting that it
is caused by mental abnormality. Thus Mr Wilson says that Tibetans
are peculiarly cold-blooded. On the other hand, Father Desideri says
they are of lively temperament. Many travellers assert an absence of
jealousy to explain it, among them Westermarck. ‘Jealousy’, he tells us,
‘is a passion of very great intensity . . . belonging to the nature of man.’

GROUP MARRIAGE (i) 137

So that ‘promiscuity is to be explained by the absence of that part of
human nature’. Westermarck has not thought it worth including in the
instances he cites the Santals who, far from possessing ‘an abnormally
feeble disposition to jealousy’ arc ‘usually frantically jealous’. Their
jealousy, however, assumes a peculiar form; they ‘often complain that
their husband’s younger brothers are carrying on intrigues with other
girls when they can get all they want at home’.

Polyandry among the ‘ Aryan ’ Hindus

It would be a mistake to suppose that the disappearance of polyandry
among the Dravidian and Mongoloid races of India was due to the
influence of the conquering ‘Indo- Aryan’ race, for the latter had the
same polyandrous institutions as the races they conquered. The ‘Aryans’
established themselves first in the Punjab, and it is there, if anywhere,
that their existing representatives are to be found — the Rajputs and the
Jats. Among them fraternal polyandry persists to the present day,
particularly in the district of Dehra Dhun, one of the chief cradles of
Hindu civilization, where the fair-skinned Rajputs and Jats predomin-
ate. Polyandry was reported as on the decline there in 1827, but m l %74
it was still described as ‘unquestionably common to this very day’ ; the
writer adding that, if a woman dies leaving an infant son, her daughter-
in-law is bound to rear the boy and marry him herself. Rajput polyandry
is strictly fraternal, only sons of the same mother by the same set of
fathers being permitted to enter into collective marriage relations ; and
it is associated with polygyny, although elsewhere other kin, including
first cousins, are admitted. The women enjoy the highest status and
freedom, being at liberty to leave their husbands and remarry provided
that the new husbands refund the wedding expenses of the old.

Among the Jats, of whom the Rajputs are the aristocratic class, the
custom appears to be universally recognized, though not always openly
acknowledged owing to Brahmanical influence. Thus of the Kanaur
district Sir Denzil Ibbetson says : ‘Polyandry is practised without dis-
grace by both Kanets and Jats; it is the rule, not the exception, for the
wife to cohabit with all the brothers.’ Among the eastern Jats it is ‘a
common thing, when women quarrel, for one to say to the other: ‘You
are so careless of your duties as not to admit your husband’s brothers to
your embraces.’ Much other evidence can be adduced ; for instance, in
the Sikh regiments of the Indian Army in the middle of the last century
it was the recognized plea for leave of absence for Rajputs and Jats that
their brothers were away from home and their wife left alone. Moham-
medan historians record that before embracing Islam many infidels of
the Punjab observed polyandry.

It has also been suggested that die Indo- Aryans adopted polyandry

T.M. — 5*

THE MOTHERS

138

from the aboriginal inhabitants or from neighbouring peoples; this is
intrinsically improbable, since individualistic marriage usages rarely
give way to communal ones, and still more because polyandry among
the Indo- Aryans of the Punjab is associated with survivals of an archaic
order of society culturally identical with that of the polyandrous
peoples of the Hindu-Kush, whence the invaders came.

As we have seen, polyandry was obligatory among the peoples of
Turkestan — that is, the ancient Baktria and Sogdiana — the main habitat
of the Indo-Aryans. Chinese annals of both the earlier and the later
Han dynasties refer repeatedly to the same customs among the Great
Get-ti, or Ye-tha. These are the same customs to which Herodotus
refers in speaking of the Massa-Getae: ‘Each man marries a woman, but
all wives are in common ; for this is the custom of the Massagetac, and
not, as the Greeks think, of the Scythians. And when a Massagetae
desires to have company with a woman, he hangs up his quiver in front
of the chariot and has intercourse with her openly/

The ancient literature of the Indo-Aryans refers everywhere to their
polyandrous institutions. In the sacred laws of the Aryans it is laid down
that ‘a bride is given to the family of the husband and not to the husband
alone’, although ‘a husband shall not make over his wife, who occupies
the position of a clanswoman, to other than to his own clansmen in order
to cause a child to be begot for himself.’

The two great national epics of the invading race, the ‘Ramayana’
and the ‘Mahabharata’ depict the marriage of the heroes as a group-
marriage. These two epics derive from sagas of the two main groups
into which the Aryan invaders appear to have been divided. Thus in the
marriage of the five sons (or brothers) of Pandu, which forms the theme
of the first book of the Mahabharata, the princes declare ‘Thy daughter,
O King, shall be the common wife of us all’, and the daughter dwells
with each of them for two days at a time. The sage Ydhishthira, in com-
mending this marriage, says: ‘Let us follow in the way that has been
traced by the illustrious of former ages ; this practice hath been estab-
lished, it is to be regarded as old and eternal.’ Similarly in the ‘Rama-
yana’. The full force of the evidence cannot be appreciated without a
survey of the entire literature. 3

The sage’s assertion that the practice is an ancient one is amply borne
out by the Vedic hymns, which refer repeatedly to plurality of husbands.
The Vedic family was constituted in the same manner as other polyan-
drous families; that is, the brothers continued to live together even after
the death of their fathers, and the eldest brother acted in all things for
the others, an arrangement almost invariably associated with polyandry.

1 The Mahabharata contains some passages expressing disapproval of group
marriage which appear to be late Brahmanical interpolations.

GROUP MARRIAGE (i) 139

The converse rule that a man would have right of access to his wife’s
sisters is also explicitly referred to . 1

Finally it must be noted that nowhere in the Vedas or in the Sutras is
there a word of condemnation of the practice, although they are meticu-
lous indeed in their moralizing. This fact negatives the argument that
polyandry was a later adoption.

It was Brahmanism which lead to the gradual suppression of primitive
marriage customs among the Hindus. The Brahmanical system was
individualistic; it was not an organization comparable with the Church of
Rome, but a collection of sacred individuals acknowledging no authority.
Hence it caused a change from a clan system to a caste system, and the
clan to which the individualistic priest belonged was of no significance
to him. The caste organization, based on the exclusive priestly caste,
broke up the clan organization based on group kinship. To treat, as is
often done, Brahmanical ideas as being ‘primitive Aryan’ usages is to
ignore the whole course of Indian history. Indeed in southern India,
remote from the contest with the aristocratic clans, the Brahmans them-
selves retained the constitution of the undivided family — the elder
brother alone being permitted to marry and his children being accounted
the offspring of the entire family. As a seventeenth-century traveller
noted: ‘It is a law with the Brahmans that one son only takes a wife, who
is common to all the brothers but to no one else.’ The learned Hindu
judge Sir T. Muttusami Aiyar comments that ‘the personal law of these
Nambudiri Brahmans’ was not the modern Hindu law but ‘the ancient
Hindu law as it is probably understood and followed about the com-
mencement of the Christian era’.

Polyandrous Marriage amongst other ‘ Aryan ’ Peoples

One way of dissociating ourselves from customs different from our own
is to suppose that the latter arc a peculiarity of an imaginary ‘Aryan’ or
Indo-German race. (Actually the term ‘Aryan’ — as Mueller, who popu-
larized it, stressed — applies solely to language.) It has been a point of
honour, especially with German scholars, to depict in the most exalted
terms the moral standards and social institutions of the supposed first-
cousins of the progenitors of their race; and in their zeal for the prin-
ciples of the Hindus of three milleniums ago they have often appeared

1 European commentators have attempted to explain such references away.
Thus Professor Hillebrandt claims that the sister dwelling with her father must
mean with her forefathers, i.e. dead. Some have even asserted flatly that ‘poly-
andry is not Vedic’. Dr Keith finds this too much, and rejects the suggestion
that the reason that the word for ‘wife’ usually occurs in the plural is that it may
refer to ‘hetairai*, as unlikely, since the Rig- Veda uses the phrase meaning ‘wife-
hood to a husband*, besides containing passages clearly referring to marriage.

THE MOTHERS

140

to lose all concern for intellectual honesty. In point of fact, the sexual
standards of inhabitants of the Hindu-Kush and the Punjab are remoter
from modern conceptions than those of any comparable people. The
only Aryans known to history, as we have seen, were the Medes, and the
Central Asiatic offshoots of the same race, to which the early invaders of
India belonged. These Aryans were polyandrous, as Strabo tells us. To
have fewer than five wives or five husbands was accounted a misfortune.

There is no reason to suppose that the conceptions of peoples speak-
ing languages related to Sanskrit have passed through phases different
from those passed through by other ethnic groups. The differences
between advanced and primitive peoples in this respect are due not to
racial characteristics but to the extent of their social evolution. Everything
goes to show that Western races have passed through the same phases as
primitive people of other races.

Evidence of sexual communism and polyandry are by no means
unknown among the peoples from whom European civilization has
derived, whether or not culturally or racially related to the Aryans. Thus
in Sparta, the community which preserved most faithfully the primitive
organization of the Hellenes, fraternal polyandry was legal. Tn Sparta’,
says Polybius, ‘several brothers had often one wife between them and
the children were brought up in common.’ It was the custom to ex-
change wives and to offer them to strangers, and, during the absence of
their spouses, Spartan women might take ‘secondary husbands’. Cir-
cumstances, such as the rituals of sexual promiscuity practised in some
Greek communities, confirm the inference that the Spartan institution
was a survival from a phase through which the Greek race as a whole had
passed . 1

The Etruscans, according to Theopompos, held their wives in com-
mon. We have already indicated that the institutions of Rome were once
very different. Those primitive Latin princes who did not know their
fathers can scarcely have been monandrous. From the most brilliant
epoch of the Roman Republic, Plutarch reports an incident which
divorced from this background, would appear utterly inexplicable. Quintus
Hortensius, a man of exalted character and a friend and admirer of the
second Cato, expressed a desire ‘to be more than a mere associate and
companion of Cato and in some manner to bring his whole family and
blood into community of kinship with him’. To this end he proposed
that Cato’s daughter, already married to Bibulus, should be lent to him
‘as noble soil for the production of children’. Lest Bibulus be averse
from the proposition, he promised to give him back his wife as soon as

1 Sexual promiscuity, of course, does not constitute collective sexual organiza-
tion ; nevertheless, it rarely survives in a society where individual sexual claims
are fully established.

GROUP MARRIAGE (i) 14I

she had borne a child. Cato was far from being offended, but refused the
request on the grounds that Bibulus would be unwilling. Hortensius
then suggested that Cato should lend him his own wife. To this Cato
could see no objection, but thought it only fair to talk the matter over
with his wife’s father. The latter entirely concurred, and Cato accordingly
lent his wife to Hortensius. Plutarch reports this with obvious embarrass-
ment, but Strabo saw the incident in its proper perspective. He refers
to it after mentioning the polyandrous customs of the Medes, adding:
‘In the same manner in our own day Cato lent his wife to Hortensius,
upon the latter’s request, following in this an ancient custom of the
Romans.’

Among the barbarians of Northern Europe, it was not unusual for
men to exchange wives. In Germany, until quite late in the Middle
Ages, it was legal for the husband, if his marriage proved without issue,
to introduce a friend to his wife, a usage which is said to have lingered
almost to our own time in some of the remoter districts. In Nordic
mythology the goddess Frigga, when her husband Odin goes on a
journey, cohabits with his brothers, like a Tlinkit lady.

Caesar states that the ancient Britons practised community of wives —
groups of ten or twelve, especially brothers, having their wives in com-
mon. Dio Cassius makes Queen Boudicca say: ‘It is over Britons that I
rule, men . . . who have even their wives and their children in common.’
Similar usages are reported of the Caledonians and border tribes : ‘They
live in tents, naked and barefooted, having wives in common and rearing
the whole progeny.’ Strabo was told that the Irish had free access to one
another’s wives and did not even bar incest. St Jerome, with theological
exaggeration, says of the northern Britons: ‘The nation of the Scots
have no individual wives, and, as if they had been reading Plato’s Republic
or wished to imitate the example of Cato, a man amongst them has no
wife of his own . . .’

There appears to be no reason to question, as has been done, the
accuracy of these statements. That they do not appear in Celtic literature
is to be expected, since all this literature has been set down in Christian
times by Christian priests. Nevertheless, the ‘Ulster Saga’ mentions
that the princess Clothru married three brothers and that her son had
thus three fathers. Multiple fatherhood is also referred to in mytho-
logical texts. In the Celtic family property was indivisible, and, unless a
son went to seek his fortune, brothers continued together, forming one
common household. The name of that household was not the house but,
significantly enough, ‘the bed’, and the Celtic family was called ‘the
common bed’. In short, there is no indication of a monogamic tradition
or disposition among the peoples who have been called ‘Aryan’, whether
in Europe or in Asia.

142

THE MOTHERS

Collective Marriage in Southern India

In southern Hindustan the survivals of collective sexual organization are
so general as to leave little doubt that this once constituted the usual
form of marriage. Group-marriage is in full force at the present day
among the tribes of the Nilgiri Hills, especially among the Todas. It has
been described as simple polyandry, but, as Gait’s account makes clear,
it is actually sororal-fraternal group-marriage. Not only are the brothers
of one family co-partners in the group-marriage of that family with
another, but tribal brothers are also admitted to the connubial group.
‘Notwithstanding these singular family arrangements’, says Major Ross
King, ‘the greatest harmony appears to prevail among all parties —
husbands, wives and lovers. The children live happily with their puta-
tive parents, equally well treated on every side.’

According to Dr Rivers, there is reason to believe that the Todas
migrated to their present habitat from the coasts of Malabar at the time
of the Aryan invasion. Hence they must be regarded as representing the
original form of organization of all the native races of Malabar and
Travancore — a conclusion supported by the reports of the older
travellers and by what we know of the usages of the populations of that
country at the present day. Thus fraternal polyandry is reported of the
Krisnavakkakars of Travancore. The Kanisans admit that fraternal
polyandry was formerly common among them. Today in Malabar
polyandry is most prevalent among the poorer classes. In former days
it existed among the noblest and wealthiest aboriginal caste of the land.
Concerning the marriage system of the Nayars, we have a considerable
number of accounts dating from the time of the first landing of the
Portuguese in India to the present day. Apart from a few misconceptions,
the numerous accounts are remarkably consistent and, in the light of
present knowledge, accurate.

As already noted, the social organization of the Nayars is one of the
purest examples of unmodified matriarchal organization, the mother-
hood, or ‘tarwad’, being composed of all the descendants of a common
ancestress in the female line and constituting the only unit of Nayar
society. These tarwads, of which there are a hundred, are classified in
order of nobility, and the strictest rule is that no Nayar woman may
associate with a man in a caste lower than her own. Apart from this
limitation, either sex is free to make whatever associations it wishes;
there is no permanent cohabitation or economic association between men
and women and no initiatory ceremony. As one Nayar says : ‘Marriage
among the Nayars is indeed pure and simple and unmixed with con-
siderations of civil rights of property — a marriage for the sake of
marriage alone. It is not an institution intended, as in more advanced

GROUP MARRIAGE (i) 143

Hindu societies, for the perpetuation of the family, but a social arrange-
ment intended for the peaceful satisfaction of the “blindest appetite” of
man.’ The association can be dissolved with the will of either party
without further ceremony. Mr Gopal Panikkar says: ‘There are two
sides to a marriage, a legal and a religious. Now, in the case of our
marriage both elements are wanting. . . . There is, in fact, no fixed rule
or custom as to marriages in Malabar. They arc terminable at the will of
either party and the law takes no notice of them/

While Nayar associations are limited by the law prohibiting the woman
to associate with an inferior caste, caste organization is not an aboriginal
institution and this must have been borrowed from the Aryans. Prob-
ably relations were originally confined by tribal law to certain inter-
marrying tarwads, a suggestion confirmed by the fact that today a Nayar
is supposed to marry his cross-cousin ; that is to say, in the same tarwad
as the other members of his own tarwad.

McLennan regarded Nayar polyandry as the type of non-fraternal
polyandry, in contrast with the Tibetan form, and the impression that
it was non-fraternal has continued among ethnological writers to the
present day. But there is every reason for thinking that this is an error.
Non-fraternal polyandry, as a traditionally established institution, is
extremely rare, if it exists at all. Instances, such as the collective mar-
riages of the Chukchi, or the polyandry still surviving in the Darjeeling
district, which are described as non-fraternal polyandry, are seen on
examination to be but degenerate forms of collective relations, in which
the husbands were originally brothers, own or tribal. The organization
of Nayar marriage has been obscured by the fact that Nambutiri
Brahmans were admitted to ‘marriage’ with Nayar women; and in the
lower castes, and wherever polyandry still survives among Nayar castes
at the present day, it is confined to the fraternal form.

Gemelli-Careri expressly states that the husbands in Nayar marriage
were brothers; ‘when one brother marries a woman,’ he states, ‘she is
common to all the others’. Other travellers, misled by the part played
by Brahmans in such unions, and also by the fact that in the Nayar social
system there were no ‘actual’ but only ‘tribal’ brothers (that is, members
of the same ‘tarwad’) have described Nayar polyandry as non-fraternal.
Since there was no ‘family’, there were no brothers in the family sense,
but only in the clan sense. That Nayar polyandry was fraternal follows
from the obligation to marry in certain ‘tarwads’ only, and seemingly
in that into which a man’s mother’s brothers have married. At the
present day, cultured Nayars, who are extremely reluctant to admit any
survival of the polyandrous customs of their ancestors, acknowledge
that a trace of it is retained in the relations between brothers. ‘The wife
of a brother’, says Mr K. M. Panikkar, ‘is looked upon as a person to

THE MOTHERS

144

whom one could openly, though not legitimately, pay court, and any
favour short of sexual relationship which she confers upon them is
allowed by public opinion. ... All brothers treat her half as a sister and
half as a wife.’

Nayar marriage was, in fact, identical with all the other institutions
of fraternal sororal polygamy, and differed only from the modern
Tibetan form in being of a more primitive type, the marrying groups
being unmodified matriarchal clans and not, as in Tibet, semi-patri-
archal families. It is significant that no instance of non-fraternal poly-
andry, as a general custom of spontaneous and independent origin, is
definitely known; and that, where such a practice is found, it appears
invariably to be derived from fraternal group-marriage. For if, as some
have supposed, polyandry were simply the outcome of various local and
accidental conditions, such, for instance as scarcity of women, there
would be no apparent reason why it should not commonly take the form
of a deliberate partnership between persons not necessarily related. But
no instance is known of polyandry having become customary except as a
modification of already existing fraternal group-marriage.

The connubial relations of Nayar women were preceded by a cere-
mony, the tying of the ‘tali’, which was performed on Nayar girls before
the age of eleven. The ‘tali’ is a small gold leaf through which a hole is
bored with the finger, and which is tied round the neck of the girl. (This
rite is general among Dravidian races.) It is often performed on batches
of girls, including infants. In one form of the usage there were as many
ritual bridegrooms as marriageable girls, and it is thus possible that the
ceremonial marriage constituted in those instances the actual marriage
of a group of girls to a group of bridegrooms . 1

The marriage customs of the Nayars have now become almost, though
not entirely, obliterated, but polyandry is, according to Mr Gopal
Panikkar, observed at the present day by the barber caste of Nayars.
Strange as it may seem, the group-marriage of the Nayars has been
strenuously denied in toro. Such repudiations are inevitable wherever
new moral values have been adopted and old usages abandoned. The
Parsee scholars attempt to repudiate the existence of the well-known
law of next-of-kin marriage, or ‘xvaetvadatha 5 , of Mazdanean religion,
an institution laid down in their Sacred Books, the practice of which for
centuries is a matter of history. The Todas indignantly repudiate the
practice of female infanticide, although it is well known that they

1 Nayars, to defend themselves against the sort of ethnical persecution to
which they have been subjected, have sometimes adduced the ‘talikettu* cere-
mony, but Sir T. Muttusami Ayer, the chairman of the Malabar Marriage
Commission, concludes in his report that ‘in relation to marriage it has no
significance save that no girl is at liberty to contract it before she goes through
the “tali-kettu” ceremony \

GROUP MARRIAGE (i) 145

practise it at the present day. In Africa, tribes mutually accuse one
another of cannibalism, while each denies the impeachment.

In conclusion it may be added that the institutions which were
general among the aboriginal races of India were, as might be expected,
common to the peoples of the same race in Ceylon. Among the Singha-
lese, fraternal polyandry was an established institution and was recog-
nized by law until the year 1859 ; it continued to be practised much later
in the interior.

Evidence of Polyandry among the Ancient Semites

Polyandrous marriage was a familiar institution among the ancient
Semites. As we know from the Arab philosopher Al-Biruni and others,
‘all the kindred have their property in common, the eldest being lord;
all have one wife, and the first that comes has access to her\ This
account is confirmed from the inscriptions of southern Arabia, which
commonly refer to a man’s ‘fathers’. These have sometimes been inter-
preted as referring to the person’s father, grandfather and great-grand-
father, but this interpretation is excluded by some inscriptions in which
the fathers are described as brothers. In one, a whole genealogy is given
in which brothers twice appear as the common fathers of their descen-
dants.

Early Babylonian records refer also to polyandrous marriage; the
important part played by the institution of levirate marriage which, in
Mosaic legislation, is expressly connected with the custom of brothers
living together in an undivided household, indicates the former preva-
lence of similar usages amongst the Hebrews. 1

1 Levirate marriage is described on pp. 153 6.

CHAPTER 11

Group Marriage (n)

I n Africa, owing to the extensive development of private property and
consequently of individualism, marriage has tended to assume the form
of personal economic, and even purely commercial, transactions.
Nevertheless, the continent presents examples of collective sexual rela-
tions. Thus, among the Banyoro of Uganda it was legitimate for a man
to have relations with the wives of the men he called brothers, i.e. his
clan fellows. A man might use his influence with his wife to make her
refrain from such action, but he could not accuse her of unfaithfulness
for so doing. A woman, however, was restricted to men of her husband’s
clan. Similarly among the Banyankole or Bahima — except that the
restriction to clan-brothers is not insisted on for cither sex. Much the
same is true of the Akamba, a Bantu tribe, but here limited to members
of the same clan or occasionally to intimate friends of another clan.
Half-brothers have a recognized right of habitual access to the wives of
their half-brothers of a corresponding age.

Among the Masai and other kindred tribes there is right of access
among members of the same marriage class, such classes being based on
age-groups of the men who passed through puberty ceremonies at the
same time. Individual marriage docs not take place until comparatively
late in life, prior to which the young men live in separate kraals and the
unmarried girls of corresponding age with them. But when individual
marriage takes place, the members of the age-grade claim priority of
intercourse with the bride. After marriage, temporary exchange of wives
is usual between members of the same age-grade. So also among the
Nandi and elsewhere.

The Herero, the great western branch of the southern Bantu, have an
elaborate institution of sexual communism which presents interesting
features. Owing to their isolation, they have preserved customs of more
primitive character than have the majority of Bantu. They are divided
into totemic clans, some strictly matriarchal, others patriarchal, and
appear to be in a state of transition from matriarchy to patriarchy.
Cross-cousin marriage is strictly observed. Sororal polygyny and the
levirate are customary. Speaking of the Ovaherero in particular, Dr
Hahn says that every man and every woman stands towards certain other
men and women in a particular relation called ‘upanga’, a word origin-

GROUP MARRIAGE (il) 147

ally meaning ‘companion*. Every man can claim access to tne wives of
his ‘upanga’ at any time, and conversely for women. These customs are
observed in all the neighbouring tribes. According to Dr Dannert,
actual brothers and all persons closely related by blood are prohibited
from entering into the relationship of ‘upanga*, though this does not
apply to the women. Though other writers contradict this, if Dr Dan-
nert is correct the sexual communism of the Herero presents a remark-
able anomaly. But this might have come about easily enough in relation
to the development of private property. The Herero are great pastor-
alists : and communism tends to be abolished wherever property becomes
of great value. It is noticeable that the prohibition regarding relatives
applies only to the men. Among the Bantu the women are never owners
of cattle; it therefore appears to refer to the economic consequences
of the relationship rather than the relationship itself.

In every part of Madagascar relations are permitted between a man
and the sisters and female cousins of his wife, and with the wives of his
brothers and cousins; conversely for women, The only forbidden rela-
tions are those between a married woman and a slave, or a stranger who
is not the guest of her husband.

Sexual Communism in New Guinea and Oceania

Sexual communism is found in northern Papua, among the Kai, and
among the southern Massims of British New Guinea. Similar institu-
tions flourish in Polynesia. Thus in Tahiti and in the Friendly Islands
the aristocratic classes were united in a brotherhood known as ‘areoi*.
A wife might sleep with fellow-members of her husband’s ‘areoi’, but
relations with non-members were accounted adulterous. Sexual promis-
cuity was practised at periodic feasts.

The Marquesan group long escaped civilizing influences, and Dr
Tautain, the French administrator there, found fraternal sororal poly-
gamy derived from an originally wider form of tribal sexual communism,
as shown by the fact that it was obligatory for the bride to be placed at
the disposal of all the tribal brothers of the husband before marriage.
Similar customs are found at the other extremity of the area inhabited
by the Polynesian race, in Hawaii, which was probably the first Poly-
nesian settlement in the Pacific. Although they are not definitely
reported from the intermediate parts, there are indications that they
were not local peculiarities of Hawaii and the Marquesas. For example,
the word ‘punalua*, now translated as ‘intimate companion’, is in
general use throughout Polynesia and its exact meaning is ‘a multi-
plicity of spouses’. Very much the same customs obtained throughout
Micronesia, polyandry being reported as common among the upper
classes in the Gilbert Islands, the Marshalls and elsewhere.

THE MOTHERS

148

Collective Marriage in Australia

As we have seen, the collective sexual organization varies; the group of
husbands may consist of actual or half-brothers or may include other
near relatives; sometimes the eldest brother is paramount; sometimes
the group consists of a limited number of clan members, or of all clan
members, or of all members of a given marriage class or age-group. And
we have seen that in organizations of the fraternal type there are signs
that they were formally associated with an organization into clans or
marriage-classes. If, as we concluded, the earliest pattern was an under-
standing between two primitive groups or clans and had no reference to
individual marriage, then the most primitive form of marriage would
consist of sexual communism between clan brothers — that is, the right
of access of all members of one group to all those of opposite sex in the
corresponding group.

Special interest therefore attaches to Australian clan organization
since there the primitive pattern has remained substantially unchanged.
Their material culture has, of course, undergone important changes,
and, as we have seen, male domination has attained its most despotic
form. Australian society exhibits several different principles of organiza-
tion which have become combined and have modified one another.
There are three basic principles of organization — totem groups, terri-
torial classifications (members of the same camping-ground) and mar-
riage classes. Each varies from one tribe to another: thus in one case the
totem is transmitted through the mother, in another through the father
and in a third the community decides to which totem a particular child
shall belong. Sometimes change of residence transfers an individual
from one territorial group to another.

This complexity of organization means a corresponding complexity
in the rules governing sexual relations, and indeed this is the sole pur-
pose of all this organization. It is therefore inappropriate to speak of
promiscuity, and these rules may properly be described as concerned
with morality, although so different from the moral rules of civilized
communities. The Australian rules are not concerned with safeguarding
the claims of individual marriage but with the prevention of incest. To
marry within the prohibited degrees would have dire consequences for
the whole tribe. Among the Dieri, the elders in their leisure lecture the
young people on the heinousness of incest, the penalty for which is
usually death. Even in cases of rape, the victim is first asked by her
assailant to what class she belongs. The totemic system is regarded as
having been instituted solely in order to distinguish between those who
may and may not have sexual relations.

Individual marriage for economic purposes is widely found in the

GROUP MARRIAGE (il) 149

present day, but is regarded simply as due to economic necessity. But
while it establishes the rights of the husband to the company, obedience
and labour of the woman, it does not regulate sexual relations. Group
marriage between members of corresponding marriage classes has long
been widespread, and was accurately described for the first time in 1832
by Scott Nind, who noted that men did not marry until after thirty, as a
rule, but that they often courted a wife while her husband was still
living. Exchange of wives among tribal brothers is ‘extremely frequent’
in Western Australia, and, as one cleric observes, ‘it may be said that
they have their wives in common’. Among the Dieri, when two brothers
are married to two sisters they probably live together in a group mar-
riage of four. In the Kunandaburi tribe intercourse constantly takes place
between ‘a group of men who are own or tribal brothers and a group of
women who are own or tribal sisters’. Among the northern tribes
fraternal polyandry is so much a matter of course that it is usually not
the husband but one of his brothers who chastises a woman guilty of
adultery.

However, Australian sexual organization differs in some respects from
the examples of fraternal polyandry earlier considered, and represents a
more primitive form. Thus it is not an organized marriage of the whole
fraternal family group, as is the marriage of the Tibetans or the Todas.
Indeed there is no such thing as a juridically-recognized family group
in Australian social organization.

Missionaries have often been misled by these institutions to suppose
them licentious aberrations. Any promiscuity which exists is a rigidly
defined group promiscuity. Thus among the four clans of the Aldolinga
tribe, a Beltare man can cohabit with any Kumare woman and a Burule
with any of the Bunanke women.

Spencer and Gillen have described some aspects of sexual relations
among the Dieri and the Urabunna tribes of Central Australia; among
the Dieri, at the great tribal gatherings in connection with puberty, a
special ceremony known as ‘Kandri’ is performed at which the names
of various couples are solemnly read out by the head of the totem group.
The various persons thus allotted to one another are not consulted, and
they are expected to cohabit immediately. These pairs are known as
‘pirrauru’, a term which appears to mean ‘moon spouses’. (In the
Warramunga tribe, it is the moon god which is supposed to have estab-
lished the marriage classes and sexual rules.) All the marriageable and
married people, even those who are already ‘pirraurus’, are allotted in
batches, and once a ‘pirrauru’ always a ‘pirrauru’. These relationships
are independent of the individual economic marriage known as ‘tippa-
malku’, and are attended with special solemnities. Every man, married
or unmarried, has one or more ‘pirrauru’ wives, and each woman may

THE MOTHERS

150

have a number of ‘pirrauru’ husbands in addition to her ‘tippa-malku*
partner. A man exercises marital rights over his ‘pirrauru* with the
formal assent of the ‘tippa-malku’ husband if the latter is present, and
as a matter of course if he is not. Similarly among the Urabunna tribe the
normal marital pairs are called ‘nupa’, but in addition each man has
certain ‘nupa’ women with whom he stands in the relation of ‘piraun-
garu\ Conversely a woman is the ‘piraungaru’ of certain other men
while being the ‘nupa’ of one particular man. In this tribe individual
marriage does not exist even in name, as far as sexual rights are con-
cerned.

The marriage organization of the Australian aborigines suggested to
Fison and Howitt the view that collective group relations must have
preceded any form of individual marriage, a view which has been
endorsed by most eminent authorities, though not by Dr Westermarck,
who is severely criticized by Spencer and Gillen for his scepticism.

The existence of group-marriage is not at issue ; the questions which
are in dispute are: (1) whether the individual association was anterior or
posterior to the sexual communism, and (2) whether sexual communism
does or did extend to the whole of the intcr-marriage classes. Dr Wester-
marck founds his hypothesis that individual marriage came first on
nothing more than the supposition that such ideas are innate in human
nature ; but, as we shall show in more detail later this rests upon a mis-
conception. What is spoken of as jealousy in primitive societies does not
refer to sexual possession. The Rev. J. Mathew bases his objection on the
ground that this would ‘reduce mankind to a state of degradation lower
than the brutes’; adding ‘if gorillas have the decency to pair off, why
may not primitive man have done the same ?’ He doubtless derives the
notion that gorillas pair from Dr Westermarck, who gave considerable
currency to that piece of misinformation at one time.

To suppose that individual marriage was replaced by sexual com-
munism is difficult, for such a trend is of course unknown in any sphere
of social development. The whole of history proves the difficulties of
checking individualistic tendencies once they have developed. Wester-
marck remarks that polygyny and polyandry are prone to be modified
towards monogamy, but elsewhere claims that the Urabunna custom
may have developed from ordinary individual marriage. It is hardly
possible for both views to be correct.

The existence of sexual communism has usually been assigned to the
difficulty of obtaining wives when there is a scarcity of women, and to
the benevolence of the older married men in remedying the distress of
the younger members by allowing them access to their own wives.
Unfortunately almost every account tells us that the scarcity is caused
chiefly by the older men monopolizing the women, and that, far from

GROUP MARRIAGE (ll) I5I

attempting to remedy the situation, they go out of their way to make
matters more difficult.

The bulk of the criticisms have been directed against Dr Howitt’s
account of the customs of the Dieri, but it seems to me that the ‘pir-
rauru’ institution constitutes but a small part of the evidence pointing
to the fundamentally collective character of aboriginal social organiza-
tion. Certain inconsistencies due to misinformation or misunderstanding
have been made much of, and it is easy to draw up a list of points
concerning the institutions of this tribe which still remain obscure, but
there are many points which are clear. For example, while a man may
have several ‘tippa-malkus’, a woman can have but one; and, in general,
the system is more beneficial for the male, as one would expect in a state
characterized by male supremacy. Mr Thomas argues that, since the
‘pirrauru’ relationship comes after a ‘tippa-malku’ marriage in the case
of the woman, it must be evolutionarily later; but since both forms of
marriage are preceded by free love, he should therefore infer that this
was the earliest state of humanity. He claims that the ‘pirrauru’ relation
cannot be a survival of unregulated relations, since it is regulated ; but,
in fact, throughout the world many forms of promiscuity arc subject to
a certain regulation, from the Nanga gatherings of Fiji to the Baby-
lonian rites of Mylitta.

It is recognized that ritual licence, which often has a semi-religious or
magical character, cannot be adduced as evidence of normal social
relations, but in practice it is often difficult to draw a hard-and-fast line
between the two phenomena. In Australian tribes, which are fragmented
into small camps which only meet at periodic gatherings, promiscuity
can never be other than ceremonial. The fact that the husband’s con-
sent must be requested, if he is present, before a ‘pirrauru’ relationship
is consummated has been thought to indicate the vested right of the
husband to dispose of the woman at his pleasure, and thus to make the
whole thing merely an instance of husbandly authority. But it is clear
that the husband never refuses his consent, and the object of asking it is
to assure him that his wife will not be permanently taken from him.
Again, the fact that a man who has access to a man’s wife offers the
latter some present has been taken as indicating that the transaction
has the character of a prostitution. Actually, so little mercenary is the
custom that the presents are commonly given away again by the hus-
band.

Turning to the second controversial aspect: though it is true that own
brothers frequently have connubial rights, nothing in Australian institu-
tions resembles the fundamentally fraternal polyandrous organizations
of the Tibetans or of the Todas. The circumstance that own brothers
are preferential co-partncrs is a consequence and not the foundation of

THE MOTHERS

152

Australian sexual collectivism. There is no sign of any evolution from
fraternal to some wider group ; on the contrary, the fraternal group and
the family itself remain an unrecognized, and indeed nameless, accretion
within the collective marriage-class organization.

The relations between members of corresponding marriage-classes
are entirely different from those between other men and women. They
are called ‘spouses’ (husband or wife) — the word ‘noa’ means spouse or
partner — but the term ‘noa,’ and corresponding terms such as ‘nupa’
among theUrabunna, apply equally to all members of the opposite sex in
corresponding marriage-classes. They are therefore frequently inter-
preted in English by the term ‘potential wife’ or ‘potential husband’.
Many of the controversies on this topic have arisen from the misconcep-
tion created by this misleading expression. Thus in England almost all
unmarried women, except the few within the prohibited degrees, are
potential wives of any unmarried man, yet for an Englishman to ravish
an unmarried girl is a serious offence. In Australia the relationship of
‘noa’ gives him the right to do this with impunity. Furthermore, a
European woman ceases to be a potential wife when she marries, but in
Australia she does not. In Australia, this relationship gives a native a
claim to sexual access to a woman, whether he contemplates marrying
her or not and whether she is married or not. A better term would be
‘facultative’ wife. When the Rev. J. Bulmer pointed out to a native of
unusual independence of mind what he called ‘the absurdity of the
native classificatory system’, the native replied that it was the Euro-
pean system which was absurd. ‘Why should I be so foolish’, he said,
‘as to call my wife’s sister ‘sister’ ? She is not my sister and she is my
wife.’

To sum up: of actual group-marriage relations, in the sense of regu-
lar, recognized and habitual sexual cohabitation, we have evidence only
in the ‘pirrauru’ and similar institutions and in the collective sexual
communism prevalent among Australian tribes in every part of the
continent. But Australian social organizations are not founded on any
system of fraternal-sororal polygamy, and their especial significance lies
in * e fact that they cannot have developed out of a system of individual
mai iage, but must, from the first, have been collective and polyandrous.
Hence there is no alternative but to conclude that the collective marriage
groups were originally co-extensive with the present intermarriage
classes, for there is a total absence of groups or relationships of any other
kind, apart from individual marriage. Dr Westermarck argues that, even
if group marriage was once common in Australia, this would not prove
that it was once common among mankind at large; and he adds that the
existence of kangaroos in Australia does not prove that there were once
kangaroos in England. But, in point of fact, no less than twenty-three

GROUP MARRIAGE (il) 153

different species of marsupial mammals have been found in the Oligo-
cene desposits of the south of England and northern France. To the
truly scientific mind the discovery of marsupials in Australia would
certainly suggest that mammals of the present day might not be un-
related. Similarly if we turn to the first account ever published of the
married usages of the British, we find Caesar remarking: ‘In their
domestic life they practise a form of community of wives, ten or twelve
men combining in groups, especially brothers with brothers and fathers
with sons/

This lengthy survey shows that, however rare collective sexual
organizations may be today, they are not as rare as supposed, and that
there is scarcely anywhere in the world where evidence of their recent
existence or actual existence is not to be found. The remarkable fact is
not that they are rare, but that they have survived at all down to the
present time. The only favouring factor common to these cases is the
comparative isolation of the peoples among whom they survive. They
are chiefly found among mountain populations, and it is not among such
populations we should look for aberrations, innovations or eccentricities.
Taken with the fact that the stranger who is made a tribal brother is
commonly regarded as entitled to sexual hospitality, these facts all con-
firm the conclusion already reached on other grounds that the regula-
tion of collective sexual relations has everywhere preceded the regula-
tion of such relations between individuals.

The Levirate

The custom that a man should marry his deceased brother’s widow,
known as the ‘levirate’, is well-nigh universal in distribution, and it is
invariably continuous with fraternal polyandry. 1 If it is, in fact, a sur-
vival of it, additional evidence is provided of the universality of the
principles which we have been discussing.

The levirate is observed with particular rigour among just those
people among whom fraternal or clan polyandry also obtains, such as the
Australian aborigines, the Tibetans, the Eskimo, the Tlinkit and others.
Moreover, just as when a man marries his wife’s sister no new ceremony
is necessary, so also when the deceased man’s brother takes his wife no
ceremony is necessary. In Ladakh, however, the woman can be relieved
from the obligation of marrying her deceased husband’s brother if she
goes through a ceremony of divorce with her deceased husband’s
corpse.

Again, the rule that younger brothers have sexual rights over the
wives of their elder brothers, but that the latter have no rights over the
wives of younger brothers, is also true of the levirate. On the other hand,
1 A bibliography of 240 titles is provided in support of this point.

THE MOTHERS

154

in a purely tribal society the widow may go indifferently to an older or a
younger brother. Among the Shuswap the widow is kept a prisoner until
she marries her husband’s brother, in the same way that a man is kept
under lock and key until he marries his deceased wife’s sister. Frequently
these two customs co-exist. In fact, among those who observe the
levirate, it would be difficult to find many instances in wffiich the custom
is not associated either with actual sororal polygyny or with its residual
form, the right or obligation to marry a deceased wife’s sister.

Nevertheless, one fundamental difference between the two situations
exists: in certain phases of society women tend to be regarded by the
men as acquired possessions, and are in fact ‘purchased’; whereas, with
rare exceptions, men are not ‘purchased’ by women. The woman ac-
quired by a family or clan is commonly regarded as permanently
acquired. As an educated Yoruba explained: ‘Women are never really
married twice. . . . Once married, they are attached for ever to the house
and family of the deceased husband; hence it is more usual for widows
to choose another husband from the same family.’ But the view that the
woman is acquired property, in which the whole family has an interest, is
so closely allied with the view which regards the brothers as having a
claim on the woman even during the first husband’s lifetime as to be
indistinguishable. As Dr Lindblom said of the Akamba, fraternal
succession is with them a sequel to the recognized rights of access during
the husband’s lifetime.

Juridic notions of property and of its transmission by inheritance are
of comparatively late emergence. Unless, therefore, we suppose that the
levirate did not come into use until such ideas had developed, we must
look for some prior social cause. The practice is found among peoples,
such as the Australians, who have no juridic idea of property. Even if
the wife is assimilated to property, such property must be regarded as
having been originally communal, like all other clan property. The
North American Indians, among whom the levirate was general, not
only have no conception of individual property but do not acquire
wives either by exchange of goods or of another female. Moreover, even
where the custom of the levirate appears to be interpreted as an inheri-
tance of property, it is found that the laws governing it are different
from those governing property. This is true even in Africa, where the
assimilation of the wife to property has been carried farthest. Thus
among the Koro of Northern Nigeria a man’s property passes to his son,
but his widow’s goes to his younger brothers. It is a very general usage
among the advanced African tribes for a man to inherit his father’s
wives, with the exception of his actual mother. Thus here the law of
inheritance, instead of enforcing the levirate, has the effect of abolishing
it. Again, we find rules regarding the disposal of widows which have no

GROUP MARRIAGE (il) 155

reference to juridic property. Thus among the Banyankole, if a man has
as many wives as he can keep, he may decline to take on his deceased
brother’s widows, and permit them to marry whomever they choose;
but he, nevertheless, retains his right of sexual access to them, and is,
in fact, expected to visit them as husband now and again. Similarly
among the Gwari, a widow is at liberty to return to her own people and
to marry whom she likes, but is under the obligation of cohabiting, if
only for three days, with the brother of her deceased husband, and any
children resulting are his. A similar custom is found among the Baluba
in the Congo and the natives of Theraka in East Africa.

Another interpretation put upon the levirate is that of the ancient
Hebrews, who regarded it as an obligation to raise seed to the brother,
or to build up his house, an interpretation also given by other people,
both savage and cultured. Among the Ossetes, while a widow is obliged
to remain unmarried if her deceased husband has left no father or
brother, yet she is not prohibited from living with other men; however,
any children which may result are considered the legitimate offspring of
her first marriage. A boy born in this way will succeed to the property at
the expense of daughters born in wedlock. The Ossetes display the
unusual but not unique feature that the widow passes, as often as not, to
the father of the deceased rather than to his brother. Thus the Ossete
father not only raises seed unto his son after the latter’s decease but
during his lifetime also. A father sometimes purchases a wife for his son
and cohabits with her until the son reaches the age of puberty. He may
even purchase a wife for the son which he has by this girl and cohabit
with her in turn.

The Ossetes are the same people whom the Romans called Alani, a
tribe of the Massagetae, among whom the fraternal household is the
foundation of the social order; thus, the levirate custom is traceable,
both in the economic and the Hebraic interpretation, to its original
source in fraternal sexual communism or polyandry.

Among the Dinkas the theory of raising seed is carried out even more
thoroughly, for if the widow is past child-bearing age it is incumbent
upon her to furnish a young girl to the brother or the nearest relative of
the deceased, and if there is no such relative she must also provide a man
to act as husband to the girl, the children still being accounted as the
offspring of the deceased husband, who with luck may go on procreating
a family in this way for more than fifty years after his death.

However, the view that the function of the levirate is to raise seed to
the husband is manifestly an interpretation subsequently invented to
explain an immemorial usage. And many who practise the levirate have no
notion of such a theory. Furthermore, it is obviously not a need to raise
seed which renders it obligatory among many peoples for the widow to

THE MOTHERS

156

wait, if her late husband’s brother is still an infant, for the latter to grow
up. Again, when a man inherits his father’s wives, the sons which he has
by them are not called his sons but his brothers. Similarly the sons of a
man by his brother’s widow do not call him father, but uncle.

In some parts of Africa, the levirate is viewed as a sort of purificatory
rite, designed to avert the anger of the deceased’s ghost. Interpretations
of the levirate cannot be seen as anything but variations of the usage of
succession to the wives of the fraternal group. This is the only explana-
tion which is applicable in all cases. It is highly improbable that a usage
so universal in distribution and so deeply rooted should owe its rise in
the first instance to the adventitious operation of a multiplicity of
diverse local causes.

CHAPTER 12

Promiscuity and Individual
Marriage

Whereas the regulation of relations between intermarriage groups is
sexual, individual marriage has its foundations in economics. In the vast
majority of uncultured societies marriage is regarded in an almost
exclusively economic light, and the changes which it has undergone had
economic causes.

In modern Europe, the sexual and economic aspects of marriage are
combined, and the former is conventionally regarded as the primary one.
This convention has been maintained even where, as in marriages of
convenience, the motive for the association is primarily economic and
social.

The origin and development of marriage have hence been discussed
by social historians almost exclusively in terms of the operation of the
sexual instinct — the exercise of personal choice, of jealousy and of
romantic love. Actually these are the products of the association rather
than its causes. To be sure, individual economic association between
sexual partners has inevitably tended to establish individual sexual
claims. These claims have brought about new restrictions on sexual
relations : the married woman tends to become prohibited to all but her
individual associate. In comparatively advanced stages of development,
infant betrothal has led, especially in the aristocratic classes, the restric-
tion on women’s sexual freedom to operate retrospectively, and so to the
demand that a bride should be a virgin.

Such restrictions did not become fully established until quite late in
the growth of advanced societies, and have contributed to the European
identification of the economic with the sexual aspect of marriage. With
us, marriage is, in theory and to a large extent in practice, the only licit
sexual relation. But in primitive society it is nothing of the sort. Far
from being a means of satisfying the sexual instincts, it is one of the chief
restrictions imposed upon them. Without a single exception, unmarried
females outside the prohibited degrees are accessible to all males. 1 If
ever a case were found where chastity was obligatory on unmarried

1 A bibliography of 422 tides is provided in support of this point.

158 THE MOTHERS

females other than as a result of the influence of some more highly
developed culture, the fact would be momentous, for it would be an
example of the appearance in mankind of a sentiment entirely absent in
animals. It would thus jeopardize the whole evolutionary conception
of human development.

Until the theory of evolution was propounded, men and animals were
regarded as having had separate origins. The accepted view was that the
primitive state of mankind was one of primal virtue — that is to say, it
accorded with the moral standards recognized in European societies,
and these sexual codes were supposed to represent the innate endow-
ments of primitive humanity; any variations were regarded as aberra-
tions due to corruption. When travellers first began to report the cus-
toms of uncultured races, they were found interesting chiefly because it
was believed that they would corroborate this theory. The science of
comparative social anthropology owes its origin largely to the interest
of the Jesuit Fathers. Father Lafitau’s Manners of the American Savages
compared with those of the Earliest Times was the first systematic book
on the subject. It was intended to uphold the theory of a primitive
moral revelation. The good Father scorns those authors who, like
Athenaeus, believed that ‘the men of the earliest times observed no
solemnity in their marriages, mixing indifferently like animals, until the
time of Kekrops, who laid down the laws of matrimony’. He attributes
this to the error in which men were sunk in the last days of paganism.
Tt appears to me evident, on the contrary, that marriage has always been
regarded by all peoples as a thing sacred and solemn, the rights of which
have been respected by even the most barbarous nations. . . . We have
seen that virginity has been honoured from the most ancient times,
consecrated in persons specially appointed to the cult of the gods and
held in regard among barbarians. That virtue could not be extended to
all persons for all the days of their life on account of the necessity of
procreating the human species; but, mankind being in that necessity,
conjugal faith has been respected, and marriage, shameful in its use, has
been subject to laws of propriety, modesty, pudicity and continence,
which are inspired by nature, upheld by reason and which the institu-
tion has preserved in the midst of barbarism.’ And he adds: T admit, of
course, that among some peoples, depravity and grossness of maimers
have at various times and in diverse places introduced abuses and even
shameful customs in this respect.’

The doctrine of organic evolution completely changed the premises
of social anthropology, and a galaxy of brilliant scholars soon placed the
social history of the human race upon a scientific basis by showing that
the organization and conceptions of primitive society differed pro-
foundly from those of modem Europe. However, the views of Father

PROMISCUITY AND INDIVIDUAL MARRIAGE 159

Lafitau have been recently revived by a Finnish writer, Edward Wester-
marck, who, taking little note of anthropological discoveries, ‘boldly
challenged the conclusions of our most esteemed writers’, and ‘arrived
at different, and sometimes diametrically opposite, conclusions ’. 1 Sup-
ported by unparalleled bibliographical industry, his views have exercised
enormous influence. If his views were substantiated, we should be
compelled to abandon scientific methods in studying the social develop-
ment of humanity. For Dr Westermarck, though he has endeavoured to
trace the institution of marriage to the apes, has not offered any examples
of regard for chastity among monkeys. Hence, there would be no alter-
native but to account for such sentiments by supposing a special revela-
tion. The hope that we may continue to employ the methods of science,
however, is strongly supported by many thousands of accounts and
statements referring to the sexual habits of uncultured races, derived
from all sorts and conditions of witnesses, many of them entirely devoid
of scientific notions, and many zealously anxious to represent the Euro-
pean conceptions of morality as universal.

As Dr Westermarck has most industriously collected such statements
as seem to support the moral theory of the last century, we cannot do
better than to examine the examples he cites.

Alleged Instances of Primitive Pre-marital Chastity

Some of Dr Westermarck’s instances refer to highly patriarchal peoples;
others to tribes which have long been Christianized. With these we are
not here concerned.

In other cases, however, he asserts pre-nuptial chastity to have been
a general usage when in reality it w^as confined to the ruling classes and
to chiefs. Thus he asserts it to have been general among the Yoruba, the
Ewe and the peoples of the Gold Coast, and, still more unwarrantably,
among the Tongans, Samoans and other Polynesians. This is pro-
foundly inaccurate. Similarly, he cites several Californian tribes in
which the wealthy class required a high bride-price for their daughters,
and therefore prevented access to them. The Californian Indians were,
however, ‘a grossly licentious race’. Among some peoples, where boys
and girls are mated as soon as they attain the age of puberty, while there
cannot be said to exist either pre-nuptial chastity or unchastity, neither
can it be said that any account is taken of purity. Again, it is no evidence
of pre-nuptial chastity to report that the married women are chaste. For
example, Comanche girls made ample use of their pre-marital freedom,
but were strictly faithful as wives. There are numerous other examples.

Again, statements to the effect that unmarried girls are chaste and

1 A. R. Wallace (who introduced Westermarck to the English public) in his
introductory note to Westermarck’s History of Human Marriage , i, ix sq.

l60 THE MOTHERS

are strictly guarded often refer only to their attitude towards strangers,
and Europeans in particular. Many North American tribes would not
even permit their women to become the legitimate wives of Europeans,
although intercourse was almost promiscuous with members of the
tribe. This was also true of many Caribbean races of the Mosquito
Coast. Although sexual relations were entirely unrestricted within the
tribe, a girl who had intercourse with a white man was killed by being
slowly whipped to death. In Benin, not even prostitutes were permitted
to consort with Europeans. In other cases, while no girl is expected to
observe chastity, neither is she expected to suffer violence, and she may
be guarded for this reason, for the dangers of such violence are very real.
In East Africa, among many tribes, a man will not pass a solitary un-
married girl without entering into sexual relations with her; if she
refused, he would probably kill her. Among the Siouan tribes of North
America, ‘a man has as little control over his passions as any wild beast’.
Among the Pima Indians, the young men actually used lassoes to catch
any stray female they might find. Among the Plains tribes, unmarried
girls, when they went out to dance, tied a rope around their thighs in
such a way as to prevent any violence on the part of the men. Such
precautions in no way prevented their liberty to have as many lovers as
they pleased.

Dr Westermarck cites the Rev. Owen Dorsey as saying that extra-
matrimonial intercourse is only practised among the Omaha by prosti-
tutes, but at the time Dorsey described the Omahas they were Christian
farmers sending their children to Sunday-school, and he warns us that
their customs were formerly quite different.

A retiring attitude in the bearing of girls is commonly regarded as
evidence of chastity by superficial observers, who imagine that women
ignorant of European moral restrictions must necessarily behave like
European prostitutes. In fact, the combination of modesty of manners
and unchastity is common. Thus, Dr Finsch says that the girls of the
Caroline Islands (who were entirely unrestrained in their sexual rela-
tions) were remarkably modest in their demeanour; and also in New
Zealand and elsewhere.

Many of Westermarck’s examples consist of statements to the effect
that it is rare to come upon pre-nuptial children, or even that a girl is
blamed or punished for having such a child. But it is an ethnological
commonplace that, in many uncultured peoples, though sexual relations
are unrestricted before marriage, pregnancy must be avoided or the
resulting children killed at birth. This rule is stringently enforced
among peoples with whom pre-nuptial intercourse is not merely per-
mitted but is obligatory, such as the Masai. Similarly of the Banyoro,
‘there was no idea of sexual union being wrong so long as there was no

PROMISCUITY AND INDIVIDUAL MARRIAGE l6l

conception, and the only risk run by the girl lay in her being discovered
to be with child/ The disposal of pre-nuptial children by infanticide or
abortion, which was very general in uncultured societies, is apt to sug-
gest to us that shame and dishonour attach to pre-nuptial motherhood ;
but infanticide is not regarded as a criminal act in the lower stages of
culture, but simply as a measure of expediency. The Australian abori-
gines destroyed most of their offspring simply because the babies were
‘too much trouble to look after’. In seasons of drought, the tribes near
Adelaide killed all new-born children and ate them. Among the Queens-
land tribes a girl’s first child was almost invariably killed. Among the
Veddahs a man is not allowed to have more than three children; all
above that number are killed, as also in the Marshall Islands. A young
married man in the Island of Rook, being asked if he had any children,
replied that he had killed them all; he was still too young, he modestly
explained, to have a family. Failure to exercise such moderation is
regarded as unpardonable improvidence. In Papua the custom was to
despise the mother of a numerous family. Among the Tshi-speaking
peoples of West Africa a woman cannot have more than nine children.
When the tenth is born, it is buried alive and the husband and wife
separate. Among the Lala of Nigeria it is customary for the father of the
bridegroom to have the use of the bride until she has conceived three
times, abortion being procured by means of a compression bandage
around the lower part of the abdomen. Children born in wedlock are
disposed of as commonly as those bom out of it. Among many North
Indian tribes a young woman’s chances of marriage were augmented by
her having had many lovers, but the unmarried women destroyed their
offspring in order to prolong their period of sexual freedom. No evi-
dence could therefore be more irrelevant as regards pre-nuptial chastity.

When all such invalid instances are eliminated, Westermarck’s
enumeration of statements concerning peoples said to enforce pre-
nuptial chastity is reduced to very moderate dimensions. In several of
these remaining instances there can be no doubt the facts are quite the
reverse of what he suggests. In others, the authorities he cites do not say
what he ascribes to them, and sometimes they say the exact opposite. Thus
he cites from Petroff a statement, ascribed to Father Veniaminoff, in
which it is asserted that, among the Aleutians, girls ‘who gave birth to
illegitimate children were to be killed for shame and hidden’. But what
Father Veniaminoff actually says is just the reverse. After stressing that
irregular relations were common, that guests shared the marital rights
of the husband, and that prior to 1825 scarcely a virgin was to be
found above the age of twelve, he adds, ‘infanticide was indeed very
rare. For down to the present day the belief is prevalent that if a girl, in
order to hide her shame, should kill her child before or after birth,

T.M. — 6

THE MOTHERS

162

countless misfortunes would be brought down upon the whole village/
All other testimonies as to the morality of the Aleutians are in agree-
ment, and they do not differ from the other Eskimo races. Yet Dr
Westermarck cites the latter also in support of his case.

He gives a list of references concerning North American tribes,
asserting that in the passages indicated ‘we read that the girls were chaste
and carefully guarded’. In the majority we read nothing of the sort, and
in some we read the exact opposite. Thus, Father Morice is accused by
Dr Westermarck of reporting this of the various D£ne tribes. What
Father Morice actually said on the subject was that the Dene had no
word for a virgin — a deficiency which becomes intelligible when we
learn that the Dene regarded sexual intercourse before puberty with
strangers as absolutely imperative. They believe that menstruation can-
not occur without it, and when missions were established amongst them
nothing astonished them more than the discovery that a virgin can
menstruate. Father Morice also refers to the account of McLean who
says that 'the lewdness of the women could not possibly be carried to a
greater excess; they are addicted to the most abominable practices,
abandoning themselves in early youth to the free indulgence of their
passions. They never marry until satiated with indulgence.’

Father Chrestien Le Clercq represented the Canadian Indians not
only as chaste but as ascetic. Men and girls resorted at night-time to one
another’s couches without any impropriety, and even preserved their
chastity for a year after marriage. This was not the view taken by other
missionaries, who untiringly denounced the usage: when one of them
preached against the custom a young Indian accused the Fathers of
desiring to gain possession of their mistresses for themselves. Father
Theodat describes the usage thus: ‘The young men have licence to
addict themselves to evil as soon as they are able, and the young girls
prostitute themselves as soon as they are capable of doing so. Even
fathers and mothers commonly act as pimps to their daughters. At
night the young women and girls run from one hut to another, and the
young men do the same, and take their pleasure where they like, without,
however, using any violence, for they rely entirely upon the will of the
woman. The husband does the same with regard to his nearest female
neighbour, and the wife with regard to her nearest male neighbour; nor
does any jealousy appear amongst them on that account, and they incur
no shame or dishonour.’ Westermarck refers to Le Clercq’s story on
the authority of Father Charlevoix, who in fact comments that ‘it
appears quite improbable.’ (He also refers to Heriot’s account which is
merely a transcription from Charlevoix and Lafitau.)

The statements adduced by Dr Westermarck in reference to the
alleged regard for pre-nuptial chastity in Africa are, when not irrelevant

PROMISCUITY AND INDIVIDUAL MARRIAGE 163

and misleading, as questionable as the foregoing. We are informed, for
instance, that in the Madai or Moru tribe the girls are carefully looked
after. But Emin Pasha, a most accurate observer, stated that the un-
married Madi girls slept in special huts for the express purpose of
allowing the young men to have access to them. Major Stigand reported
that a man is allowed to pass the night with an unmarried Madi girl if
he gives her a present of five arrows. Dr Westermarck refers to the
Dinka of the Upper Nile, concerning whom we have Captain O’Sulli-
van’s detailed study. He tells us that ‘simple seduction is not a grave
offence among the Northern Dinka’. Among the ancient inhabitants of
the Canary Islands, we are told that a woman who lost her virtue was
ostracized for life. We are, however, only told this on the authority of
the grossly inaccurate Father Abreu de Galindo, whose reports are
uniformly contradicted by every contemporary testimony. It was, on the
contrary, obligatory for every Guanche girl to lose her virtue before she
could be married.

Dr Westermarck is no happier in India where, for example, he reports,
on the authority of the Rev. Sydney Endle, that among the Kacharis,
Rabhas and Hajongs ‘sexual intercourse before marriage is rare . . .’. A
different view is taken by the authors of the last Government Report on
those tribes. Concerning the Sayeins, Westermarck reports that the
young people of both sexes are domiciled in two long houses at opposite
ends of the village, and when they may have occasion to pass each other,
they avert their gaze so that they may not see each other’s faces’. In fact,
this tribe is rigorously exogamic, none being allowed to marry except his
cousin in another village. The scrupulous avoidance rules are therefore
not directed against unchastity, but against what, by their tribal law, is
incest. When a meeting takes place between two intermarrying villages,
as on the occasion of a wedding, orgies marked by unrestrained promis-
cuity take place.

Our knowledge of the wild tribes of the Malay forests is defective,
and various edifying accounts have been circulated. Father Bourien
denies them, saying: ‘A long sojourn among erratic tribes has taught me
that from among carnal sins they only exclude one, viz. rape.’ A Russian
traveller has described the sex-relations of the Jakun as a ‘round of
temporary cohabitations regulated by chance and inclination’. Mr R. J.
Wilkinson says that the Sakai ‘leave everything to sexual passion’. These
tribes are included by Dr Westermarck in his list of peoples who show
an innate regard for chastity.

Though he has not been able to discover any example of primitive
chastity in Micronesia, he finds one in the Philippine Islands to which
he appears to attach much importance, and makes it part of the main
evidence for his thesis. While other accounts tell us that ‘the women are

THE MOTHERS

I64

extremely lewd, and even encourage their daughters to a life of un-
chastity’, Dr Westermarck cites from the German poet Chamisso, who
spent nearly a week at Manila, the statement that ‘some of the indepen-
dent tribes of the Philippines held chastity in great honour . .

The extreme licence notoriously prevalent in every part of Polynesia
is alluded to by Dr Westermarck with the characteristic concession that
‘there is said to be, or to have been, the greatest freedom before mar-
riage’. However, he gives prominence to statements calculated to convey
the impression that the ritual virginity required of the brides of chiefs in
Tonga and Samoa represented general customs. In fact, these were
exceptions. In Samoa, there was exactly one virgin per village.

In support of his assertion that ‘among many uncivilized peoples both
sexes enjoy perfect freedom previous to marriage . . . the same can
certainly not be said of the Australian aborigines while in their native
state’, he adduces a statement from Pastor Strehlow; this invests the
Aruntas with a fanatical ascetism which has escaped the notice of every
other observer, including Sir W. B. Spencer and Mr Gillen, who spent
some four years amongst them. Mr Collins, after referring to the habi-
tual practice of rape, says : ‘Even children make it a game or exercise,
and I have often, on hearing the cries of the girls with whom they were
playing, run out of my house thinking some murder was committed,
but have found the w r hole party laughing at my mistake.’

Dr Westermarck has frequently lectured or admonished Australian
scientists who have spent their lives in the study of the native races, and
has passed unfavourable judgments upon their critical standards. Had
any of them made inferences concerning the past upon such slight
evidence as Dr Westermarck produces about the present, the reproofs
which he has bestowed upon them would not be impertinent.

Effects of Contact with Europeans

The glaring discrepancy between moral theory and observed fact was
accounted for by the Jesuit Fathers by assuming that the primal inno-
cence of unsophisticated savages had been corrupted by contact with
civilization. General Denys, the Governor of New France, was per-
suaded that the morality of the Canadian Indians was due to European
contact; as an instance, he mentions that a girl found a husband more
readily if she had already borne a child, because he was then assured
that she was not barren — a preference which seems difficult to trace to
the corrupting influence of European ideas! He adds that the men do
not repudiate their wives as readily as formerly, and that polygamy is
disappearing — changes which are to be regarded, apparently, as the
effects of European corruption. Dr Currier, after a full review of the
present condition of North American tribes, concludes that ‘in nothing

PROMISCUITY AND INDIVIDUAL MARRIAGE 165

has the influence of education and Christianity been more positive and
noteworthy than in the improvement which has taken place in some
localities with regard to marriage and the sexual relations’.

This somewhat puerile theory has been revived by Dr Westermarck
who assimilates the absence of extra-connubial restrictions in primitive
societies to prostitution, and suggests that this absence is due to corrupt-
ing influence. European society has, among other things, introduced
venality and prostitution for profit, which in fact does not exist in
primitive societies in the form known to us. In Australia, wherever
white men were settled near the aborigines, wholesale prostitution of
the native women took place, but this cannot be ascribed solely or even
chiefly to the whites. Natives who are favourably disposed towards any
white man may place their women at his disposal; while many of the
women were attracted by the knowledge that the white men would
treat them with kindness.

In Polynesia, while girls freely offered themselves to the crews of the
first European ships, far from this being for profit the natives over-
whelmed their guests with presents as well.

In India, wherever premarital sexual freedom is recognized within a
tribe, it is punished when exercised with members of another tribe, in
almost every case. Among the tribes of the Chittagong Hills who prac-
tise it most uniformly, it is precisely the most inaccessible tribes which
do so most generally, and it is becoming discouraged among those in
contact with European civilization. Hence, as Mr Gait says. Dr Wester-
marck’s theory (of the contaminating effect of civilization) ‘is not in
accordance with our experience of India’.

With regard to Africa, Mr Dudley Kidd uses stronger language, say-
ing that those who believe that the Kaffir races guarded female purity
and were subsequently corrupted by Europeans, must ‘be either fools or
knaves’.

Difficulty of Distinguishing between Marriage and Other Sexual Relations

The primary purpose of marriage could not have been the satisfaction
of sexual impulses, since sexual relations (within the prescribed limits)
were freer before marriage than after. So much so that of the Angami
Nagas it is said that ‘chastity begins with marriage’, and among the
tribes of Upper Burma ‘it is claimed that unchastity after marriage does
not exist owing to their freedom of experiment before marriage’.

It has been represented that such freedom may be a crude way of
affording young people an opportunity of choosing permanent partners.
However, among many primitive peoples there is no relation between
the two. Among the Bhuiya ‘intimacy between boys and girls of the
same village does not commonly end in marriage, for which a partner

1 66

THE MOTHERS

should be sought from another village 5 . While, among the Kumbi, if
young people have had intercourse with one another they are forbidden
to marry.

As we shall see, primitives commonly undertake more permanent
relations in mature age only, usually for economic reasons and quite
irrespective of the sexual relations of their younger years. Among the
Masai the risk that pre-nuptial freedom might lead to marriage alliances
was strictly guarded against by tribal law.

Among other peoples indiscriminate unions may develop into more
stable ones, and the husband may be chosen from among pre-nuptial
lovers, as among the Trobriand Islanders. And sometimes, as among the
Igorots, when a girl has become pregnant she commonly marries the
father of the child. Among the Lolo of Upper Tonkin, an experimental
marriage is contracted by a girl spending a night with a suitor at his
house. She then returns home, and continues to live a life of sexual
freedom as before. If, after a while, she goes back to her suitor pregnant,
she is received as his wife, although he knows that he is unlikely to be
the father of the child. If she does not, the engagement lapses. In
Cambodia the parents absolutely refuse to entertain propositions from a
suitor for the hand of their daughter unless he has first seduced her;
should he suggest marriage before this, he is scorned as a fool. Similar
freedom of experiment is common among many peoples without the
fact of pregnancy constituting an obligation to marry; for instance,
among the natives of the lower Congo. As Father Merolla said : ‘These
people were accustomed to commerce with their wives for some time
before they married them, to try if they could like them; and after the
same manner the wives were to experiment their husbands . 5 Among the
Munshi of northern Nigeria ‘a boy may live with a virgin as his wife if
he gives her mother ten cloths and a pig, on the understanding that the
girl’s offspring belong to her family, and that, unless he can presently
make an equivalent exchange, he must give her up 5 .

Many Central and South American races advocated child marriage,
among them the ancient Peruvians. The agreement was binding for one
year. Previous cohabitation was regarded as so essential a preliminary
that a woman who had married without such preparation was not
regarded as respectably wedded, and was liable to have the fact thrown in
her face if the marriage turned out unsuccessfully. In spite of the efforts
of the Church and severe decrees issued by the Spanish authorities, such
customs are still regularly observed among the Catholic Indians of
Bolivia; the Arawak races of Guiana also have the same usages. Among
the ancient Egyptians a trial year was so ingrained a conception that it
persisted into Christian times and even after the Arab conquest.

We thus find every degree of transition between general pre-sexual

PROMISCUITY AND INDIVIDUAL MARRIAGE 167

relations wholly unconnected with prospective marriage, the right of
experiment or trial marriage, and recognized freedom of relations be-
tween prospective spouses. It is difficult to draw a line between such
customs. Among many peoples such customs have become less free and
open as a result of contact with higher cultures or the development of
individual proprietory claims ; they have often survived as a right to trial
marriage. But it is equally difficult to draw a line between trial marriages
and true marriages, for with many primitive peoples the latter are
scarcely more stable.

In point of fact, it is impossible to frame a definition of marriage which
will apply to all forms of the relation as found in uncultured societies,
while excluding the most casual sexual congress — as we must do if
marriage be regarded as essentially a sexual relation in accordance with
the usual European sentiments. Dr Westermarck has proposed, as a
definition, to regard marriage as ‘a more or less durable connection
between male and female, lasting beyond the mere act of propagation
till after the birth of the offspring’. But where the whole distinction, as
we conceive it, between sex relations which are, and those which are not,
marital turns precisely upon their degree of permanency, the use of such
a phrase as ‘more or less’ vitiates the definition — as we have seen with
regard to the North American Indians whose associations were in widely
varying degrees ‘more or less durable’. Among all the tribes it was
customary for a young man, when he went on a prolonged hunting
expedition, to arrange for a young woman to accompany him, both as a
sexual companion and to assist him with carrying, cooking, etc. The
woman received a liberal share of the profits, and the transaction was on
a business footing. Similarly, young men with no female relatives to
look after them would engage some young woman to perform the duties
of a wife, though both of them would remain free to visit their other
mistresses or lovers from time to time. But the relationship spoken of as
marriage among the Indians was not much more durable than these
associations. ‘The Delawares and Iroquois’, says Loskiel, ‘have seldom
marriages of long continuance, especially if there are no children
soon. . . . The family connections of Indians are commonly very exten-
sive on account of their frequently changing wives.’ The Cherokee
Iroquois ‘commonly change wives three or four times a year. . . . Mar-
riage is accounted only a temporary convenience.’ Separation takes
place without any formality. ‘Those savages are not even able to imagine
that there could be any difficulty about the matter.’ They ‘laugh at
Europeans for having only one wife, and that for life; as they consider
that the Good Spirit formed them to be happy, and not to continue
together unless their tempers and dispositions were congenial’. As will
be seen. La Hontan was scarcely exaggerating when he said that ‘what

THE MOTHERS

1 68

is spoken of as “marriage” amongst the North American Indians would,
in Europe, be spoken of as a criminal connection’.

How can we distinguish irregular intercourse and marriage in such
cases? There was no ceremony or solemnization of any kind. The
associations which turned out more permanent were not distinguished
from the more transient ones by any special contract. No distinction
can be drawn on economic grounds, for economic contributions from
both parties took place whether the association lasted a hunting-trip or
a lifetime. Lieutenant Timberlake says of the Iroquois: ‘Courtship and
all is concluded in half an hour without any other celebration, and it is as
little binding as ceremonious.’ In the union of the most permanently
married and confirmed old couple, the man acquired no more rights
over the children than the casual lover had. The children in either case
were part of the mother’s family. Father Le Jeune remarked: ‘The bond
so strong, which holds man and wife under one yoke, will be very hard
to fasten upon these savages.’

The same difficulty in drawing a distinction between marriage and
other sexual relations is found in many parts of the world. For instance,
among the Dayaks of Borneo, in the Nicobar Islands and elsewhere.
Among the aboriginal tribes of Malaya, ‘it is nothing rare to meet
individuals who have been married forty or fifty times’. Of the natives
of the Maidive Islands, we learn that it very often happens that a man
would marry and divorce the same woman three or four times in the
course of his life. Among the Ainu, as formerly among the Japanese,
there was no clear distinction in language or in usage between transient
and more durable unions. Similar reports come from such northern
tribes as the Samoyeds and Aleutians, throughout Central Asia and
from many parts of India. Sir Henry Yule says of the Khasis, ‘their
unions can hardly be honoured with the name of marriage’. Almost
identical words are used by African travellers about the Bushmen and
many others. Turning to South America, we learn that the lives of the
Guaycurus are ‘a quick succession of marriages, separations and re-
marriages, in the course of which everyone mates with everyone else,
and the same couples come together several times’. Such examples
might be greatly multiplied. As Dr Westermarck quaintly puts it : ‘There
are unions which, though legally recognized as marriages, do not
endure long enough to deserve to be so-called in the natural history
sense of the term.’ Most writers in speaking of them are, on the con-
trary, reminded of the natural history of animals.

However, there is one thing which has a consolidating effect upon
individual unions: the birth of offspring. Indeed, it is considered by
most uncultured people as constituting the consummation and estab-
lishment of the marriage relation, and a sexual association is not regarded

PROMISCUITY AND INDIVIDUAL MARRIAGE 169

as a marriage until children are actually born. Among the Todas the
expression for 'to be married’ is 'to have a son’. Among the Baila of
Rhodesia the marriage ceremony is not completed until after the birth
of the first child ; while in other cases the topic is not broached until this
stage. A woman who has not borne a child is usually treated in primitive
societies as an unmarried girl, and is called a virgin. Sometimes, and this
is true even of advanced societies, a man does not contribute to the
maintenance of his wife until she has borne a child, until when expenses
are defrayed by the bride’s father. Such usages pass imperceptibly into
the almost universal rule that the barrenness of the wife is a legitimate
ground for divorce, and for the refunding of the bride-price, if any.
Thus among the Bade of the Sahara, after the bride-price is paid, the
father builds a house by the side of his own for the couple ; if, after a
reasonable time, no child is born, the payment made is refunded and the
husband departs.

Since the association is recognized by the woman’s family only after
the birth of offspring, it tends to acquire a more durable character after
that event. But it would be incorrect to suppose that this is a general
rule. Indeed, in some cases the birth of children, far from consolidating
the association, causes its dissolution. Among the Iroquois, 'sometimes
an Indian forsakes his wife because she has a child to suckle, and marries
another whom he forsakes in her turn for the same reason’. Separation
after the birth of children was, we saw, the rule among all North
American Indians. In many parts of the world, married couples separate
if they do not get on happily together, the woman usually taking the
children with her, and sometimes she settles down with her progeny and
a new husband on the very day on which she leaves the old.

The Juridic Conception of Marriage

Even though the native tends to distinguish marriage from other
sexual associations by the fact of a child being born, nevertheless we
cannot make the formation of a permanent association the grounds of
the distinction, for little or no greater notion of permanency attaches to
unions held to constitute marriage than to others. Moreover, in all
societies which preserve their matriarchal character the birth of off-
spring does not necessitate the continued association of the parents, for
the mother’s family and not the child’s father are responsible for the
maintenance of the child. Nor does any new grouping, or family, result
from the birth of children; the father may or may not live with the
mother, and if separation takes place, he has no claim or liability in
regard to the children. Only in more advanced social phases, as a result
of somewhat elaborate transactions and indemnities, does he acquire
the right to remove any of his children from the maternal group.

THE MOTHERS

I70

In short, individual marriage cannot be based upon the formation of
a family group, since none is formed. What the birth of offspring does
is to establish, not a new group but a relationship between the husband
and the wife’s family. Thus among the Cree Indians the young husband
coming to live with his wife’s parents is treated as a stranger until his
first child is bom, whereupon he takes its name and is attached thereby
to his parents-in-law rather than to his own parents. Such usages are
very widespread, especially among peoples who have preserved a matri-
archal organization, but they still subsist among many who have long
since adopted patriarchal usages — for instance, among the Arabs, where
the custom is firmly established.

The relation thus established between a man and his wife’s family
remains unaffected when actual relations between the two are severed.
However transitory the association, this cannot in any way alter the
relation of the father to his child’s maternal family once it has been
established.

These circumstances have confused observers more concerned with
stretching primitive conceptions on the Procrustes’ bed of European
ideas than with understanding them, and have led them to strange
inconsistencies. Thus the writer who describes the utterly transient
nature of marriage among the Guaycurus of Brazil, nevertheless states
that, if a woman has a male child, she does not separate from her hus-
band until death. Yet he mentions at the same time that a woman having
a son by one husband is commonly married to another, while the father
of the boy marries another woman. The writer who makes such contra-
dictory statements obviously confuses the ‘indissoluble’ character of the
relation established between the father and the mother’s clan — the
tribes are strictly exogamous — with the supposed indissolubility of the
association between them. Perrot makes the same kind of statement
concerning North American Indians. In Samoa young men belonging
to a chief’s family used to form temporary unions with young women of
good family. The union was celebrated by exchanges of valuable
presents, and they were regarded as the most honourable unions into
which any Samoan girl could enter. After their termination the women
were debarred from contracting other alliances, and devoted themselves
to the entertainment of visitors, their function of public prostitutes in no
way detracting from their enhanced respectability. In the island of
Engano, so indissoluble is the connection established between the
husband and the wife’s family that if a man marries after his wife’s
death he must pay compensation to her family.

Of some peoples it is said that they draw no distinction between
marriage and cohabitation, or that cohabitation constitutes marriage.
But with the majority of even the most primitive peoples a distinction is

PROMISCUITY AND INDIVIDUAL MARRIAGE I7I

indeed drawn, as with ourselves, between the two relations, and the
distinctive character of marriage, with them, as with us, lies in the
establishment of a social and juridical relationship, although the nature
of that juridic foundation necessarily varies with social and cultural
conditions. Thus among the Australian aborigines, who usually give a
sister or some other female in exchange for a wife, a woman for whom
no other female has thus been given is not regarded as being ‘properly
married’. Elsewhere a woman is not regarded as legally married if the
bride-price has not been duly paid. In every instance, the sanction of
her family or immediate guardian, and sometimes of all her relatives, is
regarded as essential. In the later phases of civilization other sanctions
of a religious or moral nature may be added to these, and the perfor-
mance of the ceremony may come to be regarded as essential. For in-
stance, where a feast is regarded as essential, the man who has been
unable to afford such entertainment is regarded as not being married,
even if he cohabits with the woman for many years and has a numerous
family by her.

The distinction drawn between individual marriage and other sexual
relations is thus essentially the same in primitive societies as in our own.
In England, at the present day, a man and woman may cohabit in the
most devoted manner for fifty years and rear a family, yet, unless the
union has been legally registered, they are not married and their children
are bastards ; whereas the most transient association, or even one without
any cohabitation, constitutes, if legally established, a true and valid
marriage. The distinction does not rest upon the association or its per-
manency, or even upon the formation of a family group, but solely upon
the juridic and legal transaction. In short, in its origin individual
marriage is not rooted in any form of association between sexual part-
ners, or in any group or family resulting from such association, but even
in its most primitive and rudimentary forms it is distinguished as a
juridic relation irrespective of its stability. There is, however, a differ-
ence between the primitive and the advanced view of the distinction,
since in the latter all other relations between the sexes have come to be
regarded as illegal, or illegitimate, and subject to censure. Far from this
being the case among the North American Indians and others, the
young women freely employed abortion and infanticide in order to
avoid establishing a relationship which would have constituted a legal
bond. Again among the Line islanders and the Hawaiians, juridic
marriage served certain specialized economic purposes in the trans-
mission of landed property, and was not entered into by the majority of
the people, who had no interest in such transactions. But there is no
ground for supposing that the one sort of union was more durable than
the other. On the contrary, we are told that all were equally loose. The

THE MOTHERS

172

distinction lay purely in the juridic conditions attending the union.

A tendency must inevitably develop for a juridically-established rela-
tion to cause a depreciation in the esteem in which relations not so
established are held; the correlative of the sanctioned and legitimate
union comes in time to be an unsanctioned and illegitimate one —
though, as we shall sec, several other factors have contributed to giving
marriage the character of being the sole recognized form of sexual
association.

CHAPTER 13

Primitive Jealousy and Love

Masculine Jealousy

It has been supposed that exclusive personal claims over women
originally arose in consequence of masculine jealousy. Dr Westermarck
considers that the force of jealousy affords the strongest argument
against ancient promiscuity. However, even a cursory inquiry shows
clearly that, although instincts of sexual jealousy may be as strong in
primitive as in civilized societies, or even stronger, the claims to which
they refer differ completely from those which inspire the jealousy
of a romantic lover — so completely as to make this argument irrele-
vant.

The almost universal assumption that the sentiment corresponding
to jealousy exists in animals has already, I think, been shown to be a
misconception. Male animals will battle with rivals for the opportunity
of satisfying their sexual instincts in the same manner as they will battle
with competitors for food, but when their hunger, nutritional or sexual,
is satisfied they are indifferent. The w r ord ‘jealousy’, as usually employed,
denotes the choice of a particular individual; animal jealousy has no
reference to individual mating, but is, on the contrary, conspicuous
where such mating is absent. The natural outcome of such an attitude
to sexual satisfaction would be, in human conditions, sexual communism
and not individual association. The ‘jealousy’ manifested by men in the
lower stages of society is identical with the ‘jealousy’ exhibited by male
animals. Statements that the men of a given race of savages are ‘extremely
jealous’ are very common, but are of little value and give rise to absurd
contradictions. Thus Mr Curr asserts that among the Australian abori-
gines a husband is ‘very jealous’, yet on the next page tells us that he will
often prostitute his wife. Morenhout reports that the men of Radak, in
the Carolines, were ‘extremely jealous’; pre-nuptial licence, exchange of
wives and extreme licentiousness were, however, habitual amongst
them. Dr Westermarck informs us that jealousy is a characteristic of
the Tlinkit, the Aleuts, the Hawaiians and the Nukahivans, all these
being peoples whose sexual customs are collective. He also instances the
Eskimo. An Eskimo once told Captain Rasmussen that the only cause of
unpleasantness between himself and his wife was that she was averse to

THE MOTHERS

174

receiving other men. ‘She would/ he complained, ‘have nothing to do
with anyone but him — that was her only failing. 5 It is obvious that the
use of the same term to denote these sentiments and the passion of an
Othello arises from an unscientific confusion.

Dr Westermarck defines jealousy as ‘an angry feeling aroused by the
loss, or the fear of the loss, of the exclusive possession of an individual
who is the object of one’s sexual desire 5 . This is just what jealousy
among primitive human races is not. What the savage fears is not loss of
sexual possession but the actual loss of the woman in her sexual and,
still more, her economic aspect. The offences commonly appearing in
reports as adultery consist generally of abduction, and the anger of the
husband is completely allayed by his being presented with another
woman. Captain Tench tells us of a jovial Australian native who was
fond of expatiating upon the merits of his wife, to whom he appeared
deeply attached. Some time later, Captain Tench asked him how his
wife was getting on. ‘Oh! 5 replied the Australian, ‘she has become the
wife of Cotbee. But 5 , he added with an air of triumph, ‘I have got two
big women to compensate for her loss! 5 He was obviously the gainer.
In Northern New Guinea, if a woman runs away with a lover, her hus-
band applies to her family, who either refund the bride-price or supply
another woman. In New Britain, the men, who are stated to be ‘fiercely
jealous 5 , never have any hesitation in parting with their wives to anyone
who is prepared to refund the expenses incurred. In Samoa the abduc-
tion of a wife, we are told, was frequently the cause of tribal wars, but
mere seduction was thought very little of. Gilyak husbands take little
notice of their wives 5 infidelities, but if a woman leaves home with a
lover it is quite a different matter. Similar attitudes are found among the
Tartars, the Mishmis of Bengal and in many African tribes such as the
Kuku, the Warega of the Congo and others. The general rule is that a
woman is free to leave her husband for her lover if the bride-price is
duly refunded. No additional compensation is claimed and no grudge
borne. Among the Medge of the Congo, the seducer of the wife of a
chief is liable to be mutilated and put to death; but if he supplies two
new wives, he is neither blamed nor punished. These principles, uni-
versal in uncultured societies, obtained among our Anglo-Saxon fore-
fathers within Christian times. A law of King Aethelbert (ad 560-616)
provides: ‘If a freeman he with a freeman’s wife, let him pay for it with
his “wer-geld” and provide another wife with his own money and bring
her to the other. 5

The abduction of women, even from tribal brothers, is extremely
common in primitive societies, especially where patrilocal marriage
obtains. One of the reasons which the Australian aborigines give for their
polygamy is that an ample reserve of wives is an indispensable provision

PRIMITIVE JEALOUSY AND LOVE 175

against the constant liability to the loss of one or two. In more matri-
archal societies the woman is free to leave her husband. On the other
hand, the sanction of the husband pledges a tribal brother or guest not
to abuse the privilege by abducting the woman. Father Charlevoix
asserts that the Iroquois were extremely jealous, although they them-
selves ‘utterly denied that they were given to such eccentricity’. What
they claimed, as Hunter reports, was ‘the sole disposal of their wives ;
and although in many instances they devote them to the sensual grati-
fication of their friends without ascribing the least impropriety to the
transaction, yet they regard a voluntary indulgence of passion on their
part as an unpardonable offence’. Many North American Indians were
wont to cut or bite off the nose of a woman convicted of actually running
away, but freely lent their wives to fellow-clansmen or strangers. How-
ever, if a stranger, after having knowledge of a woman by consent of the
husband, were to repeat this without the husband making the offer a
second time, he would be killed.

Many primitive tribes are described as extremely jealous because on
the approach of strangers they hide their women. The reason is that they
are afraid lest the women should be taken away, as would certainly be
the case if the strangers were from a hostile tribe. Thus the natives of
Easter Island, when visited by Captain Cook, hid most of their women ;
but, having learned that the white men had no desire to remove them,
when visited a few years later, pressed their women upon the visitors
with annoying persistence. The wild Veddahs of Ceylon guard their
women so watchfully that it is difficult for a European, and much more
for a Sinhalese, to catch a glimpse of them. But M. Moszkowski suc-
ceeded in inducing them to let him see their women after he had pro-
duced a photograph and assured them that he possessed a wife of his
own. They are even more ‘jealous’ regarding their dogs.

Jealousy is also ascribed to Australian aborigines with as little reason.
At night, when the old men are sitting round the fire, says Mr Oldfield,
the whole neighbourhood of the camp resounds with the low whistles
which are the recognized signals of invitation of the younger men to the
women to come and join them. The older men, perfectly well aware of
the meaning of these signals, do not interrupt their conversation, and
take little notice of their wives’ irregularities.

Indifference Concerning Connubial Fidelity

Widespread evidence that little notice is taken of adultery can be ad-
duced. Throughout British Central Africa, says Sir Harry Johnston,
infidelity is treated with indifference. Referring to Brazil, Father
d’ Anchieta says that, although adultery is habitual, he never heard of an
adulterer being killed by an offended husband. They are, he says,

THE MOTHERS

17 6

absolutely indifferent to the conduct of their wives. Similar reports
come of the Conebo tribes of the upper Amazon, the Tocantin tribes of
the Maranhao region, the Guarani tribes and others.

The Aleuts, the Kamchadals, the Gilyak, the Samoyeds, as well as the
Mongol tribes of Central Asia, follow the same pattern. Among the
Kafirs of Hindu-Kush ‘cases of infidelity are extremely common’.
‘When a woman is discovered in an intrigue a great outcry is made, and
the neighbours rush to the scene with much laughter. A goat is sent for
on the spot for a peace-making feast between the gallant and the hus-
band. Of course, the neighbours also partake of the feast; the husband
and wife both look very happy and so does everyone else.’ Among the
Jats of Baluchistan it is a common saying that a tribesman who puts his
camel to graze with a Jat becomes thereby the master of the Jat’s wife.
If a stranger casts his eye on the wife, the husband disappears discreetly.
A similar attitude is found among the Aheriya of the North-western
Province, the Chamars of Bengal, the Todas of the Nilgiri Hills, who
can hardly ‘conceal their scorn for European ideas on the subject’, the
Kunnuvans of the extreme south of India and many others. Similarly
in the Pacific: jealousy is not displayed in the Marshall Islands, the
Pelews or Tahiti. Among the Maori, who took a much more serious view
of adultery than most Polynesians, ‘in general the offence can easily be
compounded for’.

Primitive Duels

In some primitive societies contests occur between the males, but they
are institutionalised and do not go very far. Thus, among the Eskimo,
if a husband and a lover quarrel they may only settle the trouble with
their fists or by wrestling, the victor taking the woman. The Comanches,
when fighting such duels, have their left arms tied together. The Slave
Indians fight their duels by pulling one another’s hair. The fight takes
place in the presence of the assembled tribe, and the belligerents take
care not to give a mortal blow. It is customary to request the opponent
to strike first. All such duels are, in fact, staged under arranged rules,
and are more in the nature of trials of strength than of mortal combat.
These regulated contests bear no resemblance to the combats of animals
for the possession of a female. (The abduction of women is a common
cause of intertribal wars, but that is an entirely different thing.)

Primitive society is characterized by every form of communistic
adjustment in regard to sexual relations, and the member of a group
who is unable to procure a female companion is assisted in every way by
other members, just as they provide him with food if he cannot obtain
any himself.

PRIMITIVE JEALOUSY AND LOVE

177

Variable Manifestations of Jealousy

Where sharing of a woman is involved, reciprocity is the essence of the
transaction. Among the Yao of Nyasaland the uneasiness caused to a
husband by his wife’s repeated misconduct is set at rest by the offender
lending his own wife to the injured husband for an equal number of
nights, or sometimes by an amicable understanding whereby they agree
to share the wife, and possibly the household expenses. In New Zealand
such a dispute is sometimes settled by both men becoming the woman’s
recognized co-husbands. Dr Hartland has shown by numerous examples
that jealousy, as we use the word, does not exist in primitive society.

In the most highly developed societies, the lover’s jealousy is aroused
even by a thought or feeling in the beloved inconsistent with the exclusive
attachment which he looks for. Indeed, even actual infidelity may lose its
sting when it is clear there was no loss of sentimental attachment. Such
sentiments arc devoid of meaning in the lower phases of human culture.
No such mental possession is either looked for or imagined. Of the Austra-
lian aborigines, Spencer and Gillen remark that ‘for a man to have unlawful
intercourse with any woman arouses a feeling which is due not so much to
jealousy as to the fact that the delinquent has infringed a tribal custom’.

There are many instances of tribes with widely different attitudes
regarding the seclusion of women. Thus, for example, the nomad
Koryak are extreme puritans. ‘Contrary to the custom of all neighbour-
ing tribes, Koryak girls must have no sexual intercourse before mar-
riage.’ Sororal polygyny is forbidden. (Mythology and traditional tales
show that these ideas are of comparatively late development.) They are
so jealous that on a mere suspicion a man will kill his wife, and on any
solid ground for suspecting her disembowels both her and the lover.
After marriage the women are obliged to dress in rags and refrain from
combing their hair or washing their faces. But the maritime branch of
the same nation go to the opposite extreme. Their customs regarding
the exchange of wives are similar to those among allied races and ‘the
caresses bestowed upon their wives are a source of gratification to the
husbands’. They are said to be in the habit of positively pestering
Russian officials, such as the postman, to lie with their wives.

In those relatively advanced societies where personal despotism has
become established, infringement of sexual claims is often revenged
with a severity which bears no relation to the sentiments with which a
woman is personally regarded. In Africa, seduction of the wife of a king
or chief, and even the mere fact of accidentally touching her person, is
sometimes punished with death. A king of Angola had a man executed
for giving a leaf of tobacco to one of the royal wives, although he had
gone no farther than to place the leaf on a stone at a respectful distance.

THE MOTHERS

178

The punishments inflicted by despots for the adultery, or suspected
adultery, of their wives are often atrocious. The ancient Peruvians
provided that a man guilty of adultery with the wives or concubines of
the Inca, or even of attempting such a crime, should be burnt alive,
together with the woman. In addition, £ his parents, sons, brothers and all
other near relatives were to be killed, and even his flocks slaughtered; his
native town or village was to be depopulated and sown with salt; the trees
were to be cut down and all the houses destroyed’. African kings gouged
out the eyes of the guilty parties and also of their relatives, and cut off the
breasts of the women; the victims were sometimes compelled to eat their
own amputated members. Such penalties are punishments for lese-majeste
rather than for adultery. Among the Niam-Niam, seduction of the wife
of a chief is punished with horrible mutilations, but among the ordinary
people a present of cloth or of beads salves the feelings of the husband.

Conceptions of what will satisfy the honour of chiefs vary curiously.
In New Caledonia a woman who has once been the bride of a chief is
debarred from ever marrying any other man, but is at liberty to have as
many lovers as she pleases, and in fact becomes what we should call a
prostitute. The same rule for the protection of the honour of chiefs
formerly obtained in the Caroline Islands and in Samoa.

The damage to the honour of a cuckold or husband constitutes an
enormous part of barbaric jealousy : it is not infidelity as such, but the
loss of reputation which is the grounds of complaint. As Father Dubois
remarks of the Brahmans, so long as the adultery is kept secret it is
regarded as a matter of small importance. ‘It is the publicity which is the
sin.’ According to an old Chinese writer, if anybody reported to a
Tungus the infidelity of his wife, the injured husband killed not only the
wife’s lover but also the informant, whom he regarded as the chief
offender. As Burckhardt remarked of the Italians, not only the husband
but the woman’s brother and her father considered themselves bound
to avenge their honour. ‘Jealousy has therefore nothing to do with those
acts; moral feeling little; and the wish to avoid ridicule is the chief
factor.’ The Statute of Tivoli excused the killing of a lover not only
when found in unlawful intercourse with a man’s wife, but also with the
offended relative’s daughter, his mother, his sister, or his daughter-in-
law. The same conception obtains in several Eastern peoples, e.g. among
the Druses of the Lebanon. But among the Moi of Indo-China, if a
woman commits adultery, while she and her paramour incur no punish-
ment, a heavy fine is imposed on the husband.

Penalties for Adultery

‘The prevalence of jealousy in the human race’, says Dr Westermarck,
‘is best shown by the punishments inflicted for adultery.’ Consideration

PRIMITIVE JEALOUSY AND LOVE 179

of such punishments gives little support to the hypothesis that indi-
vidual marriage is the oldest social institution of the human race, and
that it arose from masculine claims to proprietory rights over females.
If adultery were an offence against such a primary relation we should
expect the offence to be universally regarded with more horror
than any other social transgression. So far is this from being the case that,
even in English law, adultery is not an indictable offence. If severity of
reprobation is to be an indication of the antiquity of the institution,
then the clan must be accounted more fundamental than the family, for
breach of the law of clan exogamy — that is, the crime of incest — is
regarded with infinitely greater horror and punished more severely than
adultery.

Where adultery is not regarded with indifference, it may lead to brief
outbursts of temper, followed by an amicable settlement. On the spur of
the moment the angered husband may commit some impulsive act, such
as disfiguring the woman, but such acts are subject to social restraints.
Among the Afghan frontier tribes resort is made to European surgeons
to remedy the disfigurement produced in the fit of temper. (An Afghan
warrior, inquiring of an English surgeon the price of equipping his wife
with a new nose, was informed that it would be thirty rupees. There was
a silence, during which he was obviously weighing the situation. He
pointed out that for fifty rupees he could obtain a brand-new wife.)

In such a fit of anger a man may kill his wife or her lover. Such an act
is sometimes excused, but generally tribal law discourages leniency. In
most societies in the lower stages of culture, the man who commits
such a murder is exposed to the blood-revenge of the relatives of the
victim, as for any other murder. In more developed societies, homicide
in such a case is excused only when it can be clearly shown to have been
unpremeditated. (Among the Pasema natives, the seducer of a married
woman might be killed by the husband if found within the house, but
not outside. Similarly, in Burmese law, if a man is found in the bedroom
he may be struck, but not after he has left the bed.)

The statement that adultery is, or was at some former time, punished
with death can only be accepted with the greatest caution. There is no
well-authenticated instance in any lower culture of death being the
ordinary legal penalty for adultery, except as an act of despotism on the
part of an autocrat. In societies which have not advanced so far as to
delegate judicial authority to a ruler or council, adultery and homicide
are regarded as private torts concerning the individuals and families
affected, and not punishable by collective action. The statement there-
fore can mean nothing more than that homicide under the circumstances
is regarded as excusable; but since homicide itself is a private concern,
it is difficult to see that the statement means anything. Speaking of the

180 THE MOTHERS

East African Bantu, the Hon. C. Dundas says, ‘I am convinced that the
killing of an adulterer was not permitted under any circumstances/ As
a matter of fact, the recognized penalty for adultery among them is a
payment of ten goats. The Tuareg, when asked what they would do in
the case of their wife’s adultery, said that they would kill the culprits;
but when pressed as to whether any such incident was known to have
occurred they had nothing to say. Western ignorance has long supposed
that the Turks, upon whom it has fastened a fantastic reputation for
jealousy, punish with death the adultery of their wives. But ‘the old
story about the sacks filled with such degenerate beauties being sunk
into the river where it is deepest are the illusions of the romanticist’.
As a matter of fact, adultery among Turkish women, which is nearly as
frequent as among Western women, is usually dealt with by the obvious
prodecure of divorce. Although statements that the penalty for adultery
is death abound, there are few instances. The statement is most fre-
quently met with reference to Africa but, in point of fact, adultery
throughout Africa is usually compensated for by a fine, and often a
trifling one. Among the Ewe of Togo the penalty is two cowrie shells,
value sixpence. In Abyssinia, it is slightly cheaper: fivepence sterling.

The claim for compensation for the breach of matrimonial rights may
easily degenerate into a means of exploitation. Frequently husband and
wife entice victims into adultery, or concoct false charges and share the
proceeds. This is particularly common on the Guinea coast. The hus-
bands ‘make great gain of their wives and ’tis with this view that they
marry many, who are so faithful that when they have admitted a spark,
they immediately acquaint their husband, who directly fleeces him.
Some pretend to be unmarried and so impose on the stranger, who, as
soon as the affair is over, is undeceived by the appearance of the hus-
band, in the same manner claiming his wife as the bullies in Europe do.’
Among the Wataveta the husbands are ‘affectionate and kindly in their
family relations’, and love their wives dearly. All they ask is that if the
wife has lovers, the fine should be faithfully paid to them. Among the
Bashamma of Nigeria it is customary to pay the fine in advance. The
business details being thus disposed of, the co-respondent is free to
commit misconduct. Other offences on the part of the wife are frequently
regarded as more serious than adultery. A woman may be divorced for
witchcraft, stealing, being a bad cook or a fool, but not if she commits
adultery.

Tribal law is far from supporting claims to exclusive possession.
Among the patriarchal Australians a man who has no wife has a right
to challenge a man who has several and compel him, if he can, to sur-
render one of them. Among the Gilyak, if a woman absconds with a
lover and the husband does not succeed in bringing her back within a

PRIMITIVE JEALOUSY AND LOVE l8l

year, his claim lapses. And among the Sakai of Sumatra the husband
must achieve the return within seven days.

Far from lending countenance to seventeenth-century religious doc-
trines, as Westermarck supposes, primitive sentiments rule the theory
of basic jealousy out of court. For if a desire for individual possession of
females was really the chief factor in human society, how could indiffer-
ence to such claims be so general in lower cultures ?

Sexual Emotion

Since the eighteenth century opinions have differed much as to the
emotional forms which sexual attraction may take in primitive humanity
and as to whether savages are capable of romantic passion. Such contro-
versies could hardly have arisen had more care been devoted to elucidat-
ing the psychological nature of those emotions.

The character of such sentiments varies enormously in different
countries and periods. M. Kostylelf points out that the romantic emo-
tions depicted by Stendhal are today extremely rare, if not impossible.
Many people find it inconceivable that sentiments which play so funda-
mental a part in human life can be subject in any degree to cultural
influences. But they overlook the fact that the form which they assume
in cultured humanity is due to the fact that they are repressed. It is
precisely because they are so fundamental that they are ready to avail
themselves of any outlet offered and to assume any form which the
culture may impart. The sexual instincts are the most malleable of any
instincts. Let them be repressed, let their direct aim be denied them,
and they will soon assume unrecognizable forms, from the depths of vice
to the highest exaltations of art and religion. M. Kostyleff attributes the
noticeable decay of romantic emotion today as compared with the mid-
Victorian era to the greater physical activity of contemporary life, and in
particular to the development of sport, which serves as an outlet for
energy diverted from the sexual channel. But the chief cause of the
change appears to be the greater freedom of intercourse between the
sexes. Young men and women, brought up in seclusion from the
opposite sex, who never met except under the primmest rules, were fore-
doomed to fall into the extremes of sentimental emotion.

In European societies the sexual instincts are repressed and denied
their direct outlet at the very time when their operation is most insistent,
namely puberty. This results in various forms of nervous disturbance,
in romantic passion, in religious phenomena, in masturbation, in vice:
manifestations which are, for the most part, unknown in primitive
societies where no such repression exists. We are not concerned here
with judging such effects. They include some of the most deplorable
and some of the most exalted features of our culture. But the latter are

1 82

THE MOTHERS

no less artificial for being admirable. Psychiatrists describe love in
pathological terms, such as ‘obsession’, ‘ idee fixe\ and ‘rudimentary
paranoia.’ Even if one disagrees with psychologists in viewing romantic
sentiments as pathological, one must regard the forms which those
sentiments have assumed in European culture as, in a sense, artificial
and abnormal. The overflowing of the sexual impulse is the outcome of
the damming of those impulses. Their concentration upon one indi-
vidual results from these restrictions. The diversity of an advanced
culture extends the scope of sexual selection while restricting its opera-
tion. In short, the personal character of sexual attraction is not the cause
of monogamy but, on the contrary, its product. These conditions are
absent in primitive societies. The instincts are unrepressed; imagination
has no power over the mind of the savage.

The profound differences which exist in the relations between the
sexes in primitive societies and in our own is illustrated by the fact that
throughout those societies, the kiss is unknown. Were the fact disputable,
it would doubtless have been asserted that kissing is part of the nature
of man, and that it is impossible to believe that there existed a time
when such a spontaneous form of caress was unknown to human beings;
for exactly the same forms of reasoning apply here as in regard to the
sentiments of sexual love or of jealousy as we know them. The reason
why the kiss is unknown to primitive humanity is that preliminaries to
sexual relations are unknown. The development of emotional tenderness
is rendered impossible by the fact that no pre-nuptial state correspond-
ing to courtship exists in primitive societies.

It has been noted as remarkable that North American Indians, whose
eloquence will bear comparison with the oratory of ancient and modern
Europe, and who possess great gifts of poetic imagination, offer no
specimen of erotic literature; they have no love-songs. The same is true
of the Polynesians, who are far from being psychologically primitive,
and who have a rich and flexible language and a store of imaginative
myths and songs . 1 There are a couple of Samoan stories referring to
sexual affection, but it is post-nuptial not pre-nuptial. The Ababuaof the
Congo also have no love-songs.

Suicide among Primitive Races

The fact that a man or woman who is thwarted sometimes commits
suicide has been adduced as evidence of romantic love in primitive
races, but suicide for all sorts of frivolous reasons is common among all

1 A well-known Maori tale, the story of Hine-Moa, appears to be a solitary
instance of a Polynesian love-story ; it contains little that can be interpreted in
terms of our notions of romantic love, and would seem to illustrate rather the
‘swayamvara’ usage, of which we find traces among the Maori, than romantic
attachment.

PRIMITIVE JEALOUSY AND LOVE 183

primitive peoples. Perhaps this is because primitive man, like a child,
lives almost entirely in the present. The impulse of the moment, how-
ever trifling in itself the object may be, determines action, uncounter-
acted by any far-reaching consideration. When Jesuit missionaries re-
buked the North American Indians for not correcting their children,
they replied that if they were to do so they would probably kill them-
selves. An old Dakota woman committed suicide because her grand-
daughter had received a thrashing from her father. American squaws
kill themselves if spoken to crossly. In the Trobriands, a young man has
been known to commit suicide because his wife had smoked all his
tobacco. A native of the Gilbert Islands hanged himself because he had
been scolded by his wife. Among the Banyoro a woman will hang
herself if her husband finds fault with the dinner.

In every part of the uncultured world there is a curious disposition to
commit suicide to cause annoyance to the survivors. Of the natives of
Savage Island it is said that ‘like angry children, they are tempted to
avenge themselves by picturing the trouble that they will bring upon the
friends who have offended them 5 . Married women often kill themselves
to avenge themselves upon their husbands. Among the Goajiros, if a
man quarrels with his wife, the latter often hangs herself from revenge,
for the husband is obliged in such a case to pay for her a second time,
and the woman exults in the trick which she is playing on her spouse.
Among the negroes of the French Sudan, men commonly commit
suicide ‘to spite their wives’, a practice also prevalent on the Gold
Coast, where if a person commits suicide and, before doing so, attri-
butes the act to the conduct of another person, the latter is required to
undergo a like fate.

Often a Chuvash goes and hangs himself over his enemy’s front door,
a method of venting one’s spite not uncommon in India also. Colonel
Tod relates how four astute Brahmans committed suicide to evade the
payment of their taxes. Thereafter the tax oflicial was universally looked
upon by the population as a murderer. Suicide is often connected with the
belief that the ghost of the deceased person can plague survivors. Mr
Manning reported of the Maori that no more marked change in habit
has taken place than the decrease in suicide, which in the first years of
his residence was an almost daily occurrence. He knew a man who cut
his throat with a very blunt razor as a cure for toothache.

It would manifestly be difficult to infer from such frivolous acts of
suicide an intensity of sexual attachment. Mr Gibbs mentions that
among the western tribes of America many instances occur of young
women destroying themselves at the death of their lovers, but he is
nevertheless emphatic that there is little real affection. Herr Detzner
observes of the Melanesians and Papuans, ‘the momentary despair

THE MOTHERS

184

caused by the natural or violent death of wife or husband, the complete
lack of energy to take up the battle of life singly, alone induce those re-
sourceless people to take the step, and love for the deceased, which in
our sense of the word is unknown to them, lends no higher hallowing
significance to the wretched deed 5 .

In short, the anecdotes collected by several writers to demonstrate the
existence of romantic sentiments among primitive peoples show just the
opposite. Most of the instances put forward turn out to be suicides
because the woman was compelled to marry an old and objectionable
suitor and not because of unrequited love.

The existence of ‘elopement’ is sometimes cited as evidence of
romantic attachment among uncultured people. Such elopements are in
fact abductions; abductions are common among all Australian abori-
ginal tribes and are the customary mode of obtaining a wife among some
of them. The procedure is hardly romantic ; in describing it Dr Howitt
has recourse to the decent disguise of Latin ; before bringing home the
bride it was the custom to perform a ceremony in which the companions
of the Romeo were the chief actors : ‘Postridie in loco quodam idoneo, a
castris remoto, juvenes delecti a gente cjus abductam seriatim stupra-
verunt. Postea autem abductoris primi femina omnino habebatur.’

We shall return to the subject of elopement in considering the means
of obtaining wives in patrilocal marriage. When not a mere form or a
means of obtaining better terms from the relatives, such elopements are,
like suicide, far more often the result of the young woman’s dislike for
an elderly or repugnant suitor than a personal attachment to a particular
youth.

Primitive Love

Many testimonies to the lack of strong sexual attachment among primi-
tive people can be cited. 1 That, in uncultured societies above the lowest
levels, personal preference, and an emotional state analogous to ‘falling
in love’, do occasionally exist appears probable. But such sentiments
appear to have no depth or stability. Thwarting of individual desires
may occasionally lead to impulsive acts of suicide, but appears in general
to inflict little disappointment. Thus, speaking of the Sauteux Indians,
Grant reports : ‘They are not insensible to the charms of love, though
indeed not so subject to its empire as Europeans are in general. Here the
disappointed lover can bear the indifference of his mistress with the
calmest fortitude.’

This does not imply that primitive man is incapable of affection or
attachment. Quite the contrary. Other things being equal, there is
probably a greater disposition to whole-hearted if perhaps less deep and

1 See p. 4off.

PRIMITIVE JEALOUSY AND LOVE 185

constant, affection in primitive than in civilized man. The whole struc-
ture of primitive society rests upon such sentiments. ‘Affection with a
savage’, justly remarks Miss Kingsley, ‘is not so deeply linked with sex;
but the love between mother and child, man and man, brother and
sister, woman and woman is deep, pure and true.’ The savage can at
once become devotedly attached to an individual in whom he feels he
can confide. The admixture of such child-like affection with the sexual
impulse in man is one of the elements in sexual love; it is the manifesta-
tion of transferred paternal instincts and the true basis of the mating
instinct. But this affection is entirely distinct from erotic sentiment, and
only related to it by artificial combination. Not being associated with
the repression of sexual instincts, such affectionate attachment does not,
as with us, have the opportunity of becoming blended with it, but
remains distinct, as with those savages who are tenderly affectionate to
their wives and yet share them with any lover. The post-nuptial affec-
tion which grows out of companionship is a different sentiment from
sexual love, and not being linked with sex is compatible with the ab-
sence of jealousy. Such affection is a conspicuous feature of polyandrous
families. Tibetan families are noted for the devotion which exists
between brothers and wives, and the greatest harmony is apparent in
Toda families. Primitive man is as prone as civilized man to sensual
desire. He is equally capable of tender affection; what is unknown to
him is the intimate combination of the two. Speaking of the Arabs, an
observer as sympathetic as Burckhardt remarks : ‘The passion of love is
indeed much talked about by the inhabitants of towns, but I doubt
whether anything is meant by them more than the grossest animal
desire. 5 No Arabian love-poetry takes account of any other aspect.
Throughout the East sexual attraction is looked upon as purely physical.
Our fiction and our drama, the association of love with marriage, are
unintelligible to the Japanese.

Late Marriage

Although sexual life begins extremely early with savages the resulting
associations are transient. The decay of sexual life takes place corre-
spondingly early, and a man of thirty may be regarded as past his prime.
It is usually when the sexual instincts are thus on the wane that sexual
associations acquire greater stability and that true attachment may mani-
fest itself. In the lower stages of culture love in its most genuine form is
an attribute, not of youth but of age. As the Eskimo say: ‘Affection
comes as a result of living together. 5

The individual marriages of the Australian aborigines are an institu-
tion of advanced life. In some districts it is rare to find a married man
under forty, and one is scarcely found anywhere under thirty. The same

i86

THE MOTHERS

is true elsewhere. According to a missionary, the natives of Formosa
seldom married before the age of fifty. Tne same is true among all the
Dravidian races of Northern India. Among the northern barbarians of
Europe ‘the men did not as a rule marry until they had reached mature
age and the restless period of youth had passed’. In America, among the
Natchez, Iroquois, Hurons and Siouan tribes the usual marriage age
was from twenty-five to thirty, but there were many who finally suc-
cumbed to importunity only at forty or fifty.

The Kaffirs of South Africa did not usually marry before the age of
thirty or thirty-five, and frequently had to be pressed by the women.
The Tuareg begin to think about marriage when they are about forty.
Similar habits were general in South America.

Standards of Feminine Beauty

Primitive man’s choice of a bride is scarcely ever determined by sexual
attraction. He prefers to marry ‘the weather-beaten and hardy women
of the tribe, who are capable and hard workers’. An Indian chief,
questioned as to his notions of female charm, ‘seemed not to have made
a study of them. Their faces’, he said, ‘might be more or less handsome,
but in other respects women were all the same.’ His further remarks
are not reported, as they are said (though this is difficult to believe) to
have caused a Canadian trader to blush. In the savage, discrimination as
regards female charm is virtually confined to a preference for youth and
plumpness over emaciation and age. Monteiro says of the African Bantu
‘they are quite satisfied and content with any woman possessing even
the greatest amount of the hideous ugliness with which nature has so
bountifully supplied them’. Among the natives of Sunday Island, the
young women are utterly neglected, the older women being much sought
after, whether as wives or as concubines. In some parts of the Solomons,
the men prefer an old to a young woman, as being more experienced.
Among the Akikuyu, young boys are known to assault old and withered
hags. Among the Fuegians, in about half the unions, the women are ten
to twelve years older than the men. Among the Iroquois often a young
warrior of twenty-five is married to a woman of forty, and Eskimos
sometimes marry women old enough to be their mothers.

The primitive, indifferent to aesthetic standards, looks for purely
utilitarian qualities which may seem to our taste coarse or repulsive. In
Africa since the women are viewed as useful workers, the nearest
approximation to the male type is regarded as desirable, instead of the
accentuation of what we regard as femininity. ‘All Negro races that I
know’, says Reichard, ‘account a woman beautiful who is not constricted
at the waist, and when her body from the armpits to the hips is about the
same breadth — like a ladder.’ Ears ‘like an elephant’ are also admired. A

PRIMITIVE JEALOUSY AND LOVE 187

pendulous abdomen is accounted attractive, and an umbilical hernia is
regarded as a specially charming trait.

The chief aesthetic character which the primitive values is adiposity.
A female who is unable to rise from a sitting position without help is
generally regarded as the ideal. In Nigeria, ‘corpulence and beauty seem
to be terms nearly synonymous. A woman of even modest pretensions
must be one who cannot walk without a slave under each arm to support
her, and a perfect beauty is a load for a camel.’ Accordingly, girls in
many parts of Africa and elsewhere (for instance, the Sahara and the
Sudan) are subjected before marriage to a fattening process, and are
severely punished if they do not co-operate. Among the Tuareg, girls
of good family are entrusted at six or seven to energetic slaves who
compel them to swallow large quantities of milk-foods and flour. The
victims weep and implore to be allowed to remain ugly. In the evenings
they are rolled and vigorously massaged in order to distribute the
acquired fat uniformly and to suppress all angles and concavities. Thanks
to this regimen and to complete idleness, they are towards the age of
eighteen monstrously beautiful, being unable to rise or displace them-
selves without the aid of two vigorous slaves, and all the warriors vie for
their favours.

Such ideals seem to be general among all uncultured races. The same
fattening process is carried out for the bride of a Polynesian chief and
by the Guanches of the Canary Islands. The Patagonians, the Eskimos
and the Kirghis Tartars also appreciate corpulence. The Kaffirs justify
their admiration for obesity on the grounds that a fat woman stands a
much better chance of weathering a season of famine than a thin one.
The countless figures of hugely obese women found in deposits through-
out Mediterranean countries, which include the most ancient specimens
of plastic art, show that the same idea of beauty obtained among the
founders of European culture. It has been said that they indicate the
former widespread existence in Europe of a race akin to the Hottentots.
But this is unlikely since stcatopygous women are an artificial product,
and admiration for such forms is a universal primitive characteristic.

Most savages approve what we should regard as unsightly, namely
long, hanging breasts, which permit the women to suckle the child
which she carries on her back by throwing the breast over her shoulder.
This feature is accordingly cultivated by means of manipulations from
earliest puberty, and by use of bands to compress the base of the breast and
elongate it. Firm, upright breasts are disliked as an indication of sterility.

Economic Grounds of Sexual Selection

The marriages of the most primitive peoples are governed by the same
economic considerations which motivated the ‘ manages de convenance >

i88

THE MOTHERS

of aristocratic families in feudal Europe. The Australian aborigine, asked
why he wants a wife, replies : To fetch me wood and water and prepare
my food.’ Similar evidence can be adduced from many societies. In the
Loyalty Islands, the choice of a wife is chiefly determined by her skill as
a gardener. Among the Ainu of Japan, laziness on the part of the wife
and failure to obtain good crops are recognized as grounds for divorce.
A Singhalese, lectured on the sacredness of the marriage tie and the
wickedness of divorce, asked what he should do if, having married a
woman, it turned out that she was unskilled in cultivating rice. In the
Ao tribe of the Naga Hills, if a marriage is arranged before concluding
the alliance the couple start on a trading expedition. If the commercial
venture turns out profitable, the marriage is proceeded with. If not, it is
broken off. An Eskimo ‘marries because he requires a woman’s help to
prepare his skins, make his clothes and so forth.’ Among the North
American Indians, ‘industry and capacity for work are above all valued,
and next, fertility’. In Brazil the chief qualities valued in a wife are that
she should understand gardening and be able to brew good beer. Among
the Banyoro, ‘Marriages are . . . entered into for utilitarian and eco-
nomic reasons.’

Today economic conditions in Uganda, as in many other parts of
Africa, have entirely changed, and the men earn good wages in factories
and on Government contracts. The result is that, although much
wealthier, they tend to avoid marriage to an extent which is causing
concern to the authorities. Marriage has fallen into disuse, not because
the men cannot afford it, but because they can afford to do without it.
The economic motive for individual marriage having disappeared, there
is none other left.

To sum up: The origins'of marriage cannot be found in a'personal
claim over particular women, arising from an innate masculine jealousy.
We must therefore look elsewhere for such origins.

CHAPTER 14

Selecting a Husband and
Acquiring a Wife

Marriage Proposals by Women

In primitive society, it is commoner for women to choose individual
men than for men to choose individual women. Thus among the tribes of
Queensland, ‘it is never usual, it appears, for the young man to make the
first advances to a young woman of his own tribe. The “gin” has the
acknowledged right of showing her partiality for a particular person.
We could not learn that the poor fellow had any right to refuse.’ In the
Torres Straits young men arc lectured on the impropriety of proposing
marriage to a girl. Even when the man has other wives, it is the woman
who proposes. And even after the girl has declared her affection ‘any
forward conduct on the part of the young man would have been regarded
as bad form’. Indeed, it was proper to ignore such proposals until they
had been repeated several times, to avoid doubt as to the lady’s inten-
tions.

Female initiative in selection is reported from Papua to India, from
northern Melanesia to South America. Similar usages appear to have
existed among the Slavs and ancient Irish, while love-letters from an-
cient Egypt tell the same tale. Where, as among the North American
Indians, it was customary for marriages to be negotiated among the
respective mothers, it was normally from the girl’s mother, not the
man’s, that the first move came. Among the Radeh, a primitive people
of South Indo-China, the young lady who has taken a fancy to a youth,
after inquiries as to whether her suit will be welcome, establishes herself
in the youth’s house and attempts to seduce him. Should she become
pregnant within a year, she can remove her conquest to her own home
without further formality — but if not, she must pay a bridegroom-price
in cloth or cattle.

Such usages are far too widely distributed to be regarded as exceptions
or aberrations. They are, as we might expect, observed in ummodifiedly
matriarchal and matrilocal societies. Where male domination and a
patriarchal social order have become fully established, the men naturally
conduct the matrimonial negotiations, and often with little regard for

THE MOTHERS

190

the woman’s wishes. For a woman to make any advances comes to be
regarded as immodest. Thus in Australia, it is from Queensland, where
the position of the women is distinctly less debased than is the rule with
the aborigines, and where matrilocal marriage usages appear to have
survived later, that feminine initiative in courting has been reported, as
also from the Kurnai tribes, which have preserved an unusually primi-
tive form of marriage organization.

Likewise in Melanesia courting by the woman is found in the New
Britain and New Ireland groups, as also among the Melanesians of New
Guinea, whereas it would be out of the question in the Solomons, in
New Caledonia or in Fiji, where brutal male domination is established.
Where the usage is found in a patriarchally organized community, it is
probably a survival from times when women had a right of choice;
while the deference which is often paid to the inclination of the young
women, and the right of refusal which they often possess where marriage
negotiations are conducted by parents or go-between, is probably also a
legacy from such earlier times.

The reason for the woman’s original right of choice is clear. In the
conditions of primitive matriarchal societies the woman is far more
interested in marrying than is the man. The economic value of a woman’s
labour is confined to the household in which she works; the man can
take advantage of it only if he removes her to a household of his own.
But if all marriage associations were originally matrilocal and a woman
never left her group, she would be devoid of economic value to her mate.
The product of a man’s labour, on the other hand, especially when it
consists chiefly of game, can be equally well distributed to his own
household group or to that of his sexual mate. In such conditions,
therefore, the economic inducement to association between sexual
partners operates on the woman alone; it is she, and the group to which
she belongs, who are gainers by the accession of a new provider of food
and a new protector.

It is biologically irrelevant to adduce the ‘passive’ character of female
germ-cells and the ‘active’ character of male germ-cells in support of a
general social law that the male instincts must be the more active factor
in bringing about a mating association. The relevant biological fact is
that wherever animal mating takes place and there is a more-or-less
prolonged association of the sexes, it invariably does so in relation to the
requirements of the female and not to those of the male.

Love of Adornment greater in Primitive Males than in Females

Darwin’s theory rested upon biological facts which everywhere show
the selection of males by the females, not the reverse. In the animal
kingdom it is the male which displays colours and ornaments; the

SELECTING A HUSBAND AND ACQUIRING A WIFE 191

female is invariably inconspicuous. Similarly in most primitive societies :
the coquetry and love of adornment commonly assumed to be innate in
all ‘daughters of Eve 5 appear to be absent or rudimentary in primitive
womankind. Thus in Australia, ‘adornments are almost entirely mono-
polized by the men: females are content with their natural charms’. In
the Gambier Islands the women not only ‘have no ornaments of any
kind’, but they ‘appeared quite indifferent to the beads and trinkets
which we offered them’.

Throughout the American continent the same difference is conspicu-
ous. With all the tribes of the South American continent ‘the use of
ornaments and trinkets is almost confined to the men’. Thus among the
Muras the men display a wealth of ornaments, while the women are
absolutely naked. Young Choroti men spend hours over their toilet
before going to a dance, while the young women trust to their natural
charms. Fuegian men have their hair dressed and brushed by the women ;
similarly among the Andamanese islanders.

In the Congo ‘toilet and luxuries of dress are the attribute of the
stronger sex, and the costume of the women is generally more simple’.
The same is true in East Africa; the men affect the most elaborate
fashions in hairdressing. The Mashukulumba have cone-shaped, taper-
ing headdresses, sometimes a yard long; the women shave off their hair,
and present their future husbands with their own locks to serve as
padding for their extravagant coiffures. The Syrian traveller, Barde-
sanes, observed the same among the natives of Central Asia near
Baktria. ‘The Gelan women’, he wrote, ‘neither perfume themselves nor
wear dyed garments, but are all barefooted, although the Gelan men
adorn themselves with soft clothing and various colours, and wear gold
ornaments and perfume themselves, and this not from any effeminacy in
other respects, for they are brave and very warlike, and much given to
hunting.’

Women 9 s Choice determined by Economic Considerations

Primitive woman’s choice of mate is determined by practical considera-
tions. Thus, among the Sea Dayaks the women ‘generally regard mar-
riage as a means of obtaining a man to work for them’ and a ‘woman will
often separate from her husband simply because he is lazy’. ‘They cling
to us’, said an Eskimo, ‘because we give them food and clothing.’ When
a hunter is sick, his wife goes to another man. Amongst all the North
American tribes skill in hunting and prowess in war is the chief recom-
mendation in a prospective husband. Similar reports come from all over
the world. Among the tribes of Brazil, a warrior who has distinguished
himself is overwhelmed with marriage offers. Among the Nilotic negroes
‘a man with a reputation as a clever hunter can take his pick any day of

THE MOTHERS

192

all the girls in the district, whereas a youth who cannot shoot straight
will find difficulty in getting a wife at all 5 . Of several communities it is
expressly stated that a marriage lasts only so long as the man continues
to provide adequately.

In the higher stages of culture, where economic pressure is less, other
factors may operate. Tuareg ladies, who besides owning most of the
property are superior in education and refinement to the majority of the
men, are said to show a preference for somewhat effeminate men and
beaux raconteurs. But such tastes are not to be found in primitive
societies.

To demonstrate ability to support a wife is accordingly the indispens-
able preliminary to marriage. In the lower stages of culture, a man
invariably pays court by presenting specimens of his efficiency as a
hunter or proofs of his prowess as a warrior. Among the North American
Indians, ‘when a young man wishes to marry a squaw, he sends her a
quarter of venison with the message : ‘I can furnish you at all times with
the game necessary for your food 5 . Similar usages appear to have been
at one time customary in Scotland: the Caledonian hero Duchomar
approaches the lady of his choice by sending her some venison. In
hunting communities no youth may contemplate matrimony unless he
has killed some animal. Among the Koyukuhotana of Alaska a youth
who has not killed a deer is thought to be incapable of begetting children.

In warlike tribes it is usually a condition of marriage that a man
should have proved his value by killing an enemy. Among the Yoruba
‘whether the killing was done in fair fight or in the form of a murder did
not matter . 5 The rules of hospitality were conscientiously observed, but
a guest, after he had left the house, was often waylaid and his throat cut,
thus enabling his host, as a fully qualified murderer, to enter the
honourable state of matrimony. Among the wild tribes of Formosa a
man was not allowed to marry until he had murdered a Chinaman. Of
the Dayaks of the West Coast of Borneo Heer Francis writes : ‘He who
succeeded in bringing back a head, no matter how obtained, was
immediately received as a distinguished member of the community and
had a free choice of all the girls of his village . 5 With them, as also with
the Alfurs of Ceram, of Minahassa, and of Sumatra, marriage was not
possible unless such a trophy was forthcoming. So also among the Naga
tribes of Assam, the Guaycurus and some tribes of the Amazon.

Tests of Endurance at Initiation and Marriage Ceremonies

In nearly all primitive communities, before a youth can contemplate
marriage, he must take part in certain ceremonies, generally referred to
as ‘rites of initiation 5 or ‘puberty 5 ceremonies. These rites vary in differ-
ent parts of the world, and generally contain magical or quasi-religious

SELECTING A HUSBAND AND ACQUIRING A WIFE 193

elements. But they agree substantially in one respect: the candidates are
subjected to ordeals in which their valour and powers of endurance are
tested, affording an opportunity to judge their qualifications as hunters
and warriors.

Such tests are commonly of the utmost severity. Thus in the Pueblo
tribes the candidates were repeatedly flagellated with bundles of yucca,
or ‘Spanish bayonet’. Among the tribes of British Columbia the would-
be warriors ran long distances over rough ground until their feet bled ;
dug large pits to test their arm strength ; cut one another’s chests, arms
and legs with knives, and had the tips of their fingers slit open. Dry
iir-needles were placed on their hands, arms, legs and chest, set alight
and allowed to burn to ash. Anyone who could not endure the pain was
ridiculed.

The refinements of torture endured by all candidates among the tribes
of the Plains surpassed in ingenuity anything devised by Spanish
Inquisitors. After a four days’ fast and abstinence from drink, the young
fellows had the muscles of their shoulders or breast transfixed with
scalping knives jagged to a saw-edge in order to render the operation
more painful; wooden skewers were then inserted through the wounds
and ropes fastened to them, by which candidates were hoisted to the
ceiling. Other skewers were similarly inserted in their arms and legs, with
shields, buffalo skulls and other weights suspended from them. The
victims were then twisted round and round so as to wind up the ropes
and cause them to whirl rapidly as the ropes untwisted. This was but
the first part of the ordeal. As soon as they had sufficiently recovered,
they ran races before the assembled tribe, dragging after them heavy
weights fastened by skewers to their bodies. All this had to be borne
without showing any sign of pain; and the young fanatics frequently
devised fresh refinements of torture of their own accord.

The famous ‘huskanaw’ ordeal to which boys were subjected at
puberty among the tribes of Virginia was, like some other rites of initia-
tion, calculated to promote the subjection of the youth to the authority
and power of the old men; but it was declared that the test ‘hardens
them ever after to the fatigues of war, hunting and all manner of hard-
ships to which their way of living exposes them.’ Death sometimes
resulted.

Among the Carib races of South America ferocious ants are often
employed to test the powers of endurance of candidates. Thus among
the Macusi of Guiana, the young men, besides thorough flagellation and
the infliction of deep wounds on their persons, have receptacles con-
taining ants applied to various parts of their body, or are sewn up in a
hammock filled with them. If the candidate utters a cry, he is not
permitted to marry. In addition, he must clear a space of forest, and
t.m. — 7

THE MOTHERS

194

bring as much game and fish as possible to show that he is able to sup-
port a family.

A peculiar trial to which the Mura lads were subjected was that they
were compelled to drink an enormous quantity of fermented liquor; the
women next administered voluminous enemas until their intestines were
so distended that the abdomens were as tight as drums. They then had
to perform violent exercises. Many succumbed to the test, and are said
to have burst like a shell in the middle of the assembly.

In Africa, among the Kaffirs and Bechuanas, the candidates are
prevented from sleeping and are compelled to take violent exercise,
running and dancing till they drop from exhaustion. They are scourged
mercilessly, and bear the marks all their lives. In the old days many of
the boys died. If they want meat, they must steal it, and should a thief
be discovered, he is beaten unmercifully for his clumsiness. The young
Spartans, it will be remembered, were treated in exactly the same man-
ner at the time of the manhood ordeals.

Circumcision is an indispensable preliminary to marriage, and is said
by the Congo tribes to make the boys strong. The manner in which they
bear the operation is a test of the boy’s quality. Among the Orang Balik
Papan of Eastern Borneo circumcision is performed immediately before
marriage, as also among the Arabs of Djezan, where it is done in a
particularly brutal manner, in the presence of the young man’s intended
bride ‘If he betrays by any groan or gesture, or by the least contraction
of the muscles of the face, the horrible pain which he feels, the bride
declares that she does not wish to have a girl for a husband.’

The Australian initiation ceremonies consist of cutting with knives,
hitting with the ‘nullah-nullah’, tearing the hair, burning the flesh,
fighting with warriors and, wonderful to relate, delivering orations. If a
candidate has the good fortune to kill or seriously wound an enemy, his
claims are at once recognized and he is admitted as a warrior.

In some parts of Nyasaland and of East Africa dances are the only
initiation ceremonials, and are at the same time the usual avenue to
marriage. The same is true of Melanesian New Guinea, though it would
appear that the obtaining of a human head was formerly requisite. The
dances are, however, extremely strenuous. Speaking of the Gualola of
California, Mr Powers remarks : ‘The amount of dancing which they can
endure for ten or fifteen days together, day and night, is astonishing,
when we remember that the manner of dancing practised by the men is
terribly hard work.’

In Polynesia, where exogamous clans and marriage-classes have dis-
appeared, the conditions of life are exceptionally easy; elaborate man-
hood ceremonies no longer exist, but are represented by the operation
of tattooing. The operation is extremely painful and protracted; the

SELECTING A HUSBAND AND ACQUIRING A WIFE 195

victim is held down by four or five young women, who sing to encourage
him. When the tattoo is healed a great dance is held, ‘when the admira-
tion of the fair sex is unsparingly displayed’. In more primitive societies
and among warrior tribes, the tattooing was but a certificate of prowess.
Thus among the Natchez and other Mississippi tribes the youths are
‘tattooed on the nose only until they have killed some enemy and have
brought back his scalp when they have the right to have themselves
tattooed elsewhere’.

Manifestly in their practical aspects the ‘manhood ceremonies’ of
uncultured peoples are designed to afford a test of the strength, courage
and endurance of young men, and of their efficiency as hunters and
warriors. It might be supposed that these tests are imposed primarily in
view of the community’s interest in the quality of its hunters and its
warriors, but this does not fit the facts. For, if so, one would expect that
uncultured peoples would also seek to produce efficient hunters and
warriors by a regular course of training, but there is usually very little
such training, and what there is often follows, instead of preceding,
the tests.

There is one institution only for which successful performance of
initiation rites is indispensable — marriage is absolutely forbidden to a
youth before he has been duly certified as having attained to the status of
hunter or warrior. Indeed, the ceremony itself is sometimes spoken of as
‘marriage’. Sometimes one and the same ceremony constitutes both the
‘initiation’ and the marriage.

Sometimes the initiation tests are applied individually as part of the
marriage preliminaries or of the marriage ceremony itself. As we have
seen, in many countries the bridegroom is soundly thrashed by the
relatives of the bride. This is part of the rituals of violence which mark
many marriage ceremonies ; but it appears in many cases to constitute a
test or ordeal in addition. Among the higher Hindu castes of the Punjab
the bridegroom, in his progress towards the bride’s home, passes
between a double row of damsels and women armed with reeds, who
strike him as he passes; the blows are quite harmless, but the ritual is
doubtless a relic of a severer ordeal.

Frequently, such tests of endurance are applied in the presence of the
proposed bride, or even by the women themselves. In Dongola it is the
girl herself who tests the mettle of her suitors. She sits down between
two of them, holding long knives in each hand, and slowly presses them
into their thighs. The man who bears the torture best is chosen as her
husband. In other instances, the women are excluded from such cere-
monies, but there is reason to believe that this was not originally the
case. Although women are, in general, quite willing to defer to the
opinion of their male relatives in this matter, often they themselves are

THE MOTHERS

196

the judges. The procedure thus bears a close resemblance to the way in
which, in ancient India, girls of noble birth selected a husband. Relics
of these ‘swayamvara’ customs are still to be found among the Rajputs.
Such marriage competitions were also much in vogue in archaic Greece,
where, as in India, the bride was allotted to the victor in an archery
competition, a foot-race or perhaps a bullfight. So also in mediaeval
Europe. As late as the eighteenth century in County Derry, Ireland, it
was the custom for the suitor of a marriageable maiden to compete for
her hand in sports, which were generally arranged by the authorities as
a Sunday entertainment.

Similar contests were widespread in South and Central America.
Thus among the Chavantcs ‘he who can carry a heavy log over the
longest distance, or excel in running, or cast a spear farthest, bears the
bride home’.

Such organized contests are found among the rudest as among the
most highly cultured tribes of the American continent. ‘The most com-
mon gateway to sexual intercourse east of the Rockies’, says Father
Morice, ‘was wrestling. Two young men would publicly wrestle for the
possession of a maiden, and the same took place in connection with any
married women as well.’

The Zulus, like most tribes of warriors, affect great contempt for
women, but there are indications that things were once otherwise.
Although women are now ‘purchased’, if there are a number of com-
peting suitors, the right of choice lies with the women. In practice, the
observance is now mostly formal. Each suitor, dressed in his best dancing
attire, comes up for inspection, squatting in front of the lady’s hut. If she
is satisfied, he is instructed to come again the next morning and exhibit
his paces in the cattle-fold before a large concourse of people.

Among the Chukchi at the present day, women sometimes choose as
their husband the victor in an athletic contest or a foot-race in the same
manner as did the daughters of Anaetus and of Danaus. Among all the
Turki and Mongol peoples of Central Asia the bride-race takes place on
horseback. Among the tribes of the Malay Peninsula ‘the bride-elect
darts off, au galop , into the forest. . . .’ A chase ensues, during which,
should her inamorato return unsuccessful, he is met with jeers, and the
match is declared off. Sometimes the bride-race takes place in canoes.
The Aieta, a negrito race of the Philippines, have a similar custom. If
the suitor finds the bride and returns with her before sunset, the couple
is considered legally married.

Marriage by Service

The view that the primary function of such tests and ceremonies was
originally to enable women to choose the ablest husband is confirmed by

SELECTING A HUSBAND AND ACQUIRING A WIFE I97

the fact that the same element of testing is present in ‘marriage by
service’.

In primitive matriarchal societies where no personal property exists,
the sole economic contribution a man can make is to contribute his
services. In comparatively advanced societies, the man may, in consider-
ation of his devoting his labour for a period to the service of his wife’s
family, be permitted subsequently to remove her to a home of his own.
But this period of service is far from being simply a contribution which
may be commuted by some other form of payment. Thus, among the
Tupi tribes of Brazil, whose marriages were in any case permanently
matrilocal, the period of ‘service’, or probation, commonly took a com-
petitive form. When a girl had a number of suitors, all would ‘serve’ for
her for two or three years. They would work for her father, dig his garden,
cut wood for him, fish, hunt and supply the household. The most
effective provider was chosen as the husband.

That such service is not merely a form of payment is evidenced by
the persistence of the usage even where the man is in a position to make,
or actually makes, such a payment. Thus among the Awok of northern
Nigeria a man pays a bride-price for his bride but, in addition, he must
work for a time on the farm of his prospective father-in-law. Among the
Koryak personal service cannot be commuted; even a wealthy suitor,
the owner of large flocks, must serve many years as herdsman to his
father-in-law. The way in which he is treated reveals the object of this.
He is not treated as w r ell as an ordinary servant, but is subjected to
indignities, and unproductive labours are imposed on him. Mr Jocbel-
son remarks : ‘The principal thought is not his usefulness, but the hard
and humilating trials to which he is subjected.’ The Chukchi suitor
likewise, no matter how wealthy, is compelled to serve for his first bride
— but probations are not required in respect of her sisters, even if he has
begotten children during his probation. He carries burdens, hauls heavy
sledges, mends broken utensils. He must please the girl’s father, her
elder brother and other male members of the family.

If the Koryak or Chukchi suitor does not come up to the standard
required, he may be dismissed without reward or explanation, even
after serving five or ten years. Among the Yukaghir, according to their
own account, ‘the period of service is intended only to test the young
man’s ability to work. The bridegroom is required to be a good hunter
and fisherman, and capable of doing everything necessary in the house-
hold.’ In fact, an able suitor is permitted to take his bride after a
reduced period, his qualifications having been sufficiently demonstrated.
The tests imposed upon him have often no direct economic value. Thus
the prospective father-in-law will go into the woods and fell a huge
tree. The suitor must then drag it to his home, as in the ‘log tests’ of the

THE MOTHERS

198

South American Indians. He usually ends by dropping the tree-trunk
on the top of his father-in-law’s ‘yurta’, and the delighted old gentleman
exclaims, as his home comes tumbling down about his ears : ‘This is a
good man, he will be able to support us.’

All matrilocal marriage is ‘marriage by service’, and the evidence is,
I believe, conclusive that this has everywhere been the most ancient
form of individual marriage contract. For wherever the two usages exist
conjointly, matrilocal marriage is either known to have formerly been
the general practice, or is definitely preferred by the family of the
woman, or is required in the form of a period of probation, whether or
not a payment is made. Where the custom of acquiring a wife by the
payment of a bride-price has come into general use, marriage by service
may persist as an alternative to such payment, and it then appears to be
a substitute for it. But it is, I think, clear in all such instances that the
payment has come into use as a commutation of such service and not
the reverse.

Marriage by Purchase

‘Marriage by purchase’ is by far the most widespread mode of
acquiring a wife, and is found both in barbaric and advanced cultures.
The ancient Semites called the bride ‘she who has been paid for’.
Hosea tells us that he paid fifty shekels for one of his wives. Wife-
purchase was the universal rule in Vedic India, among the Tartars and
Mongols and among the archaic Greeks. Among European peoples in
Christian times the payment of a bride-price was normal. The essential
features survive to the present day in southern Spain, where the bride-
groom on betrothal deposits all his savings with the bride’s family, who
use the money to buy equipment for the new home — the bed and linen,
however, being provided by the bride’s parents.

Nevertheless, only in comparatively few barbaric societies does the
transaction have the character which the phrase suggests to our ears.
Such transactions obviously cannot be assimilated to a commercial
traffic where commerce does not exist. There is no notion of barter in
native Australia; materials and objects unprocurable locally are ob-
tained by exchange of gifts, without any conception of exacting equiva-
lent values or of getting the better of the bargain. Throughout North
and South America the notion of traffic was equally rudimentary. (In
European days shell-money, or ‘wampun’, and beaver skins became a
standardized basis of exchange, but it is more than doubtful whether
‘wampun’ was so used in pre-Columbian times.) In New Zealand, as
throughout Polynesia, ‘buying and selling for a price as practised by us
was unknown’.

Where the conception of private property is undeveloped, it is a

SELECTING A HUSBAND AND ACQUIRING A WIFE I99

universal rule that if a thing is desired it is asked for and accepted with-
out any sense of obligation, and given, even if valued, without any
expectation of adequate return. At their tribal meetings the Australian
aborigines exchange gifts; they do so, as they say, in order, ‘to make
friends’. In the very lowest phases of culture it is customary for a man
to present gifts to the relatives of the woman whom he wishes to marry ;
thus, for example, among the Hill Damaras the bride-gift consists of a
bunch of onions and some striped mice. The Caroline Islanders give the
girl’s father some bananas or some fish.

In pre-Columbian days there was no ‘marriage by purchase’, in the
proper sense of the term, in any part of the American continent. The
Pacific tribes of California and British Columbia held that a substantial
marriage-gift was essential ; a system of aristocratic castes exists in those
tribes, and the object of the bride-gift was to serve as a guarantee of the
rank and wealth of the bridegroom, whose family must make a return
present of at least equal value. Where the presenting of substantial gifts
by the bridegroom has become customary among other American tribes,
the same rule is observed; as for example among the tribes of the Ore-
gon. The Fuegians have been described as ‘buying’ their wives, because
they present otter skins to her relatives; but the wife, on the other hand,
brings to her husband a canoe and a set of harpoons, a much more
valuable contribution. Manifestly, these are not mercenary transactions.
Far from constituting a sale of the woman, they are a guarantee of the
social standing of her suitor and of his ability to provide for her. In
later times, such guarantees may easily have excited the covetousness of
the girl’s family. Among the Navahos and the Apaches a considerable
number of horses was required as a bride-price for a young woman ; but
traffic in horses was not an original habit of these Indians, for they had none.

Similarly, in Polynesia ‘purchase’ was unknown in regard to marriage
or any other transaction. Yet very valuable presents were exchanged in
Samoa between the respective families of persons of noble family. In
Nakahive the bride’s family was loaded with presents, but they were
immediately given over to be looted by the people. In Raratonga the
gifts were redistributed among the givers after the wedding.

In the most primitive cultural stages, the purchase of a wife is in any
case impossible because the men possess no fundable property and
therefore no purchasing power. The primitive hunter who joins the
social group of his wife or ‘serves’ for her contributes the product of his
labour, and this may be regarded as a form of payment. In more
advanced cultural stages he owns possessions which may enable him to
commute his obligation to serve his wife’s group permanently. ‘Mar-
riage by purchase’ is thus evolved from the ‘mitigated form of slavery’
of the serving husband.

200

THE MOTHERS

Only when a man became an owner of domesticated cattle, his first
form of real property, was he in a position to commute his service by a
payment. Marriage by purchase in the proper sense is accordingly not
found at all in Australia, Melanesia, Polynesia or in America, where no
domesticated cattle, and consequently no man-owned wealth, existed.
It is in pastoral societies, or in societies which have passed through
pastoral stages, in Africa, Asia and Europe, that the purchasing power
of the bride-gift has developed.

In the vast majority of instances such a payment is not regarded as
commercial, nor is it so in fact. The ‘purchased’ wife may be a princess
and the husband nothing. The princesses of the royal house of Ashanti,
for example, took a succession of husbands, who ‘purchased’ them.
They were dismissed after they had been completely depleted of their
substance. What the complacency of Europeans has represented as the
‘purchase of women’ is thus not even an indication of female subjection.
The Greeks were embarrassed by the fact that their ancestors ‘bought’
their wives, but the ‘purchased’ wives of the archaic age enjoyed a far
higher status than the wretched dowried wives of later times.

Even where ‘marriage by purchase’ has assumed its crudest form, it is
far from being regarded by the women as an indignity. The sum paid is
a source of pride; the fact of payment is the criterion of the legitimacy
of the union. In Australia a woman who has not been exchanged for
another, or for whom nothing has been given, is socially degraded, and
is regarded as we should regard a prostitute ; in West Africa and many
other places likewise. Yakut women do not think European women quite
respectable, since they can be had for nothing, and even offer a dowry to
a man as an inducement to marry them. Where, as in South Africa, the
natives have been Christianized and have consequently been persuaded
not to pay for their wives, this has brought about an enormous increase
of immorality and loss of self-respect on the part of the women; con-
jugal unhappiness and desertion have greatly increased.

Even in Christian Europe the payment of the bride-price was once
regarded as the main condition of the legality of the marriage. In the
early Middle Ages marriages in which the husband had made no pay-
ment were regarded as illegitimate. We have, for example, a series of
Merovingian legal documents in which a man goes to great trouble to
obtain a court order establishing his children as legitimate, although he
has not paid the proper purchase money for his wife. In Denmark when,
owing to the influence of Christianity, men were becoming lax in this.
King Frotho decreed that no one should be permitted to marry a woman
unless he had paid for her.

To be sure, the acquisition of a wife by paying for her, although not
originally an act of purchase, has, by an inevitable abuse, at times

SELECTING A HUSBAND AND ACQUIRING A WIFE 201

assumed that character. Thus in Sumatra, where marriage by purchase
is a late innovation, a man in some parts of the country may, by paying
double the usual bride-price, acquire complete rights over the woman,
and may even resell her, hire her or pawn her. In East Africa, among the
Mkamba, a wife ‘is bought and sold, and may even be traded as a piece
of goods’. Similarly, in Chinese Tibet a daughter may be sold over and
over again; the speculation may thus be repeated as often as ten times.
The Samoyeds also ‘commerce with their wives’; and Georgi knew a
Tartar who had disposed of eight successive wives at a profit, with their
entire concurrence. (This remarkable business-man was at the time
engaged in courting his ninth bride.)

Examples of wives being regarded as an article of traffic may be found
nearer home. ‘It is much to be lamented’, stated Mr Grafton in a peti-
tion to Queen Elizabeth, ‘that wards are bought and sold as commonly
as beasts.’ Numerous instances of wife-sale occurred in England in the
nineteenth century.

Such mercenary instances are in reality exceptional, considering the
wide distribution of the usage, and are found chiefly where the social
organization has long been subverted by the development of slave traffic,
and where the bride-price has consequently tended to become identified
with the purchase money paid for a slave. That the people themselves
are frequently aware that such a mercenary view is an abuse is indicated
by the efforts which are often made to regulate the bride-price. In
Uganda a universal rate was fixed as bride-price for all women, whether
peasant girls or princesses, amounting to thirteen shillings and four-
pence.

In Africa, above all, ‘marriage by purchase’ has developed into a
mercenary bargain, but it is a corruption of comparatively recent origin,
and the corruption is by no means complete. To take only one example,
among the Zulus, ‘marriage by purchase’ is perhaps as crude a trans-
action as anywhere, but they ‘never consider it as a sale’; the term ‘to
buy’ does not apply to it, and the husband has no right to resell his wife.
‘The practice of making an express bargain’, says Mr Shooter, ‘can
hardly be said to have prevailed thirty years ago.’

The bride-price is sometimes sentimentally represented as the ‘price
of virginity’. Primitively it has no reference whatever to the right of
sexual access to the woman and refers solely to the right to remove her
and her children from her parental home to her husband’s. In Indonesia,
where the native custom of matrilocal marriage and the more recently
introduced usage of ‘marriage by purchase’ commonly exist side by side,
the latter confers on the husband the right to remove his wife and her
children; but he is quite at liberty to marry the woman without any
bride-price if he is content to take up his abode in her home. So also in

T.M. — 7*

202

THE MOTHERS

Tonkin, Cambodia, and many other places. Frequently, when the
required bride-price is paid in instalments, the wife stays in her own
home until payment is completed, and her children do not belong to her
husband’s, but to her parent’s family — as among the Kafirs of Hindu-
Kush, in Timorlaut, in the New Hebrides, among the Mishmis of
Assam, and the Wakamba, the Makaranaga, the Kuku and other African
peoples.

However, in firmly established matriarchal societies it is sometimes
no easy matter for the men, even when they possess the power to offer a
substantial bride-price, to break through the immemorial privilege of
matrilocal marriage, and they resort to strange devices. Since in matri-
archal law a man’s property passes to his sister’s children, not to his son,
the men in several parts of Africa have adopted the plan of purchasing
slave-girls, and of making over their property to their children by those
concubines. Thus it comes about that those children whom we should
call ‘legitimate’ do not inherit from their father, while those whom we
should call ‘illegitimate’ are the only lawful heirs.

Marriage by Capture

One mode of obtaining wives which would apparently ensure female
subjugation is the capture of women from neighbouring tribes or groups.
McLennan and Lord Avebury held that this was the original mode of
obtaining a wife; the former regarded it as the cause of the law of
exogamy, and thus of the incest prohibition. If that were so, it would
constitute the source of all marriage regulations.

Certainly, the forcible capture of women has occurred in every part of
the world at all epochs in the course of warfare. It prevailed among the
warlike tribes of central and northern Asia; and among the Germans,
Scandinavians and Slavs. The archaic Greeks and Semites constantly
captured women. Nevertheless, there is not, so far as I am aware, any
clear instance of this being the habitual method of obtaining a wife, any
more than it was among the Greeks or Hebrews. Such a practice would,
indeed, involve the tribe concerned in perpetual warfare, for the capture
of even one of its women is invariably regarded by primitive tribes as a
casus belli . The resulting mutual extermination must usually check the
practice ; in fact, we are told that among the Bakyiga, the practice ‘led to
such fierce fighting that the clan gave up the practice . . . while in
Australia the tribal elders discouraged such a dangerous practice.

In South America, the prevalence of the practice is doubtless due to
the fragmentation of the native races into numerous tribelets which are
constantly at war, a state brought about by the slave-hunting expedi-
tions of the Dutch and Portuguese. In Africa, it was prevalent where the
slave-trade reached its greatest development, as in Nyasaland and

SELECTING A HUSBAND AND ACQUIRING A WIFE 203

British Central Africa — but the slave-trade is of relatively recent deve-
lopment in Africa. No North American tribe ever made a practice of
obtaining wives by capture.

The term ‘marriage by capture* also covers three other practices : the
abduction of single women against their will; elopement with the
woman’s connivance, but not the parents’; and cases where the whole
proceeding is more or less fictitious. Such fictitious abductions are very
common, and much of the so-called evidence for wife-capture rests
upon such simulated violence. Thus among the Kurnai, when, at the
end of an inter-tribal festival, a young man elopes with a woman, her
relatives make a great show of resentment ; but it is understood that a
suitable arrangement will be made, and their wrath is merely intended
to ensure this. The man may have to meet the relatives in a formal com-
bat, in which care is taken to inflict no serious injury, and which may
reduce itself to tapping him on the head with a boomerang. Sometimes
the abduction does not take place until all arrangements with the bride’s
parents have been completed, though the bride herself may not have
been consulted.

Mock fights between the relatives of the bride and those of the bride-
groom, the former pretending to prevent her removal from her parental
home, are of almost universal prevalence. They are found in Australia,
in Melanesia, in New Guinea, in Polynesia, in Indonesia, in India and
other parts of Asia, and in Africa. Such fictitious combats at weddings
are even more conspicuous in the higher phases of culture, and are a
feature of customs among the country people in every part of Europe.

But there are serious difficulties in the way of interpreting these
usages as reminiscences of real raids. In the first place, it is not easy to
imagine why the memory of such supposed abductions and raids should
have been so generally perpetuated by innumerable peoples as a cere-
monial observance. The descendants of men who carried off brides in
victorious raids might well celebrate their ancestors’ deeds of valour,
but there appears no reason why the defeated parties should be anxious
to commemorate their humiliation. And in such instances the opposition
of the bride’s people is, in general, a far more conspicuous feature than
the assaults of the bridegroom’s friends. Some of the forms which those
simulated captures assume are quite irreconcilable with the proposed
interpretation. Thus, among several peoples the ‘capture’ of the bride
is effected by women; this is the case in Greenland, among the Bedawi
of the Sinai Peninsula and in the island of Nias. In Kashmir it is not the
bride’s house which is closed against the bridegroom’s party, but his
own; his sister barricades the door when the bride is brought home, and
is only induced to admit her on payment of a compensation for receiving
a strange woman into the household.

THE MOTHERS

204

Perhaps the strongest ground for this view is the inadequacy of the
alternative interpretations offered. Such usages are frequently inter-
preted as conventional displays of modesty or coyness on the part of the
bride or of her friends. Thus, when among the Eskimo of Greenland a
man removes his bride to his own home, it is etiquette for her to offer
the most violent resistance, and for the man to drag her, screaming and
struggling, to his dwelling. We are told that this is ‘lest she should lose
her reputation for modesty*. But since ‘it would be difficult to find a
people more cynical and more devoid of shame 5 , it is hard to accept this.
Again, among the Kamchadals, the bridegroom, to establish his right
to the bride, is obliged to undress her and touch her vulva in spite of
every obstacle. The woman is dressed for the occasion in many layers
of leather gowns and pantaloons securely sewn on her and made fast by
a multitude of straps, so that she looks ‘like a stuffed figure 5 ; moreover,
the bridegroom’s efforts are violently resisted by the elderly females of
the family.

The suggestion that this procedure is inspired by a desire to make a
display of the chastity and modesty of the bride can scarcely be recon-
ciled with the fact that the Kamchadal bridegroom has a right to
reproach the bride’s mother for her negligence should he happen to
find his bride still a virgin, nor with the circumstance that the ceremony
is performed after the man and woman have cohabited for several years.

It has also been suggested that these displays of pretended violence
are magic intended to avert the malice of evil spirits. To be sure, almost
every act amongst uncultured peoples contains provisions to avert ‘bad
luck’ ; but that the desire to avert the malice of spirits should so often
have taken the form of a simulated conflict between the families of the
bride and the bridegroom is too much of a coincidence. Still less is it
intelligible that the necessity for such magical safeguards should cease as
soon as the due compensation has been paid.

Among the Banyoro there is no simulated violence in the customary
wedding usages, but the traditional procedure includes precautions
against such violence. If simulated violence be supposed to avert the
wrath of evil spirits, it is difficult to see how precautions against the
occurrence of such violence can effect the same purpose. Compared with
such defective interpretations, the explanation that simulated opposition
is an echo of real opposition is far more satisfactory. The only objection
to the latter interpretation is that there is no evidence that forcible
capture ever was the usual and normal mode of obtaining wives among
any people.

The bride-price (as distinct from the customary presents out of which
it developed) acquires its importance only where the woman is removed
from her home; it is the price of that removal, not merely of access to

SELECTING A HUSBAND AND ACQUIRING A WIFE 205

her. Where marriage remains matrilocal there is no such development
of the bride-price; there are also no ceremonies of resistance. Both are
unknown in North America, where matrilocal marriage is the rule. In
Indonesia, where matrilocal and patrilocal marriage are almost every-
where found side by side, no bride-price and iro rituals of capture are
found except where the bride is removed. The Biblical account of the
marriage of Jacob offers a typical instance of ‘marriage by capture’.
Jacob runs away with his wives, and is hotly pursued by their incensed
father and his kinsmen, who bitterly reproach him for the rape. But the
capturing husband and the captured wives had been formally married
for over fourteen years and had a family.

There is an even more potent reason for the display of resistance than
the desire to obtain as high a bride-price as possible. The removal of the
bride from her home to that of her husband constitutes a breach, of the
oldest usages of primitive marriage. Some peoples will not be induced
by any consideration to allow their daughters to leave home and to
follow their husbands. The Brazilian Indians, when one of their
daughters was forcibly taken away by a European, followed her en masse
and yielded themselves as slaves to the abductor rather than part with
their women. Every breach of established usage, more especially of
marriage usages, is humiliating if not actually wicked. Where patrilocal
marriage customs are supplanting matrilocal marriage, proud aristo-
cratic families refuse to adopt the change, and insist upon their sons-in-
law joining their daughters in their own homes. The relatives who have
been induced by economic considerations to yield to the man’s desire
to remove the woman, are bound to ‘save their face’ by the fiction that
they arc submitting to compulsion.

One of the most familiar usages which have suggested an attenuated
survival of ‘marriage by capture’ is the widespread practice of lifting
the bride over the threshold of her husband’s house. It was observed
by the Romans, and is still found in modern Greece; it obtains in India,
in China, in Java, in Palestine, in Egypt, in Algeria and in various parts
of Europe, including England and Scotland. The interpretation of the
usage has given rise to a good deal of discussion. Plutarch though that
it recalled a time when husbands, ‘taking their wives by force, brought
them to their house’, a view which has been followed by many modem
writers. Others (e.g. Frazer) have sought its explanation in superstitious
ideas connected with the threshold as a place of particular danger and
ill-luck, but it is not obvious why a bride should be the only person to be
protected against such dangers.

It appears to me that the most natural underlying idea is that the
entrance of the bride into her husband’s home is the final act in her
transfer from her own home to that of her husband. It is a common

206

THE MOTHERS

custom to carry the bride the whole way from the one to the other.
Among the tribes of British Central Africa and Nyasaland the whole
wedding ceremony is spoken of as ‘entering the house’. Among the
Baholoholo of the Congo the bride’s entry into the man’s hut is re-
garded as constitutinjpjie consummation of the marriage.

Where patrilocal marriage has supplanted the older matrilocal usage,
the consummation of the transaction is in fact the crossing of the
threshold of the husband’s dwelling; it seems quite unnecessary to seek
in magic or other ideas a more recondite explanation of the usage.

The view taken here coincides essentially with the most obvious and
the most general interpretation of those widespread customs which are
commonly included under the designation of ‘marriage by capture’; but
the conclusion to which we are led as regards their bearing upon social
history is the exact opposite. Such usages are not evidence of the former
prevalence of a custom of procuring wives by capture, but of the uni-
versal distribution of matrilocal marriage.

Patrilocal marriage is a violation of the time-honoured order, and
must be excused by a show of yielding to force. But this change has not,
in general, been brought about by mere force (save in the rudest stages
of social evolution), but by economic conditions which women themselves
have been the chief agents in creating.

CHAPTER 15

The Social Evolution of
Monogamic Marriage

Let us now consider the question: is it true, as Westermarck and others
have claimed, that marriage was originally monogamous and that poly-
gamy represents a decadent form of marriage ?

The development of agriculture in its higher forms, and the conse-
quent establishment of a regular food supply was the great turning-
point in human history. It marks the boundary between civilization and
uncultured ‘primitive’ states, and decisively affected the sexual constitu-
tion of human society. For, so long as men possessed no fundable
wealth, marriage tended to remain matrilocal and the social order matri-
archal, except where male dominance was established by brute force. It
was the domestication of animals which first placed economic power in
the hands of men, since animals appertain to the hunter. This power
was commonly used to buy off the claims of women, and of their fami-
lies, to the services of husbands; thus patriarchal society with patrilocal
marriage became established among pastoral peoples. On the other
hand, where agriculture, which was the province of women, developed
on an important scale without any intervening pastoral stage, the matri-
archal order has often persisted, and has even become accentuated as in
North America. The matriarchal character of society has also been
preserved among many African peoples who have remained chiefly
agricultural, e.g. in Egypt where pastoral property was never very
important.

Elsewhere agriculture has reached the highly developed stage only
after passing through a long pastoral phase, as with the ‘Aryans’ of
India and the Semites of Western Asia, who were driven by the desicca-
tion of their pastoral lands towards the great alluvial plains, the granaries
of the world. Among the Semitic nations of Western Asia women
retained many relics of their primitive influence — the code of Ham-
murabi shows countless provisions protecting the status of women;
women could own property, conduct business and plead in court. Yet
Babylonian society contrasts sharply with that of Egypt in that ‘the man
is more important than the woman, the father than the mother, the

THE MOTHERS

208

husband than the wife’. Assyrian pictorial art, unlike that of Egypt,
scarcely ever represents a woman. It seems that the development of
agriculture in societies originally pastoral, instead of increasing the
power of the earth-cultivating woman, accentuate the existing supre-
macy of the owners’ flocks and gave rise to the most pronounced types
of patriarchal society.

In the poorer, more broken lands of Europe there existed neither a
fully-developed pastoral society nor large-scale agriculture. The men
never became rich enough to purchase Oriental harems. The land,
broken up into small holdings, long remained in the hands of the women
who had formerly tilled it. Hence the man came as a suitor to the woman.,
through whom alone he could enter into possession of the land.

Where agricultural civilizations evolved without any preceding pastoral
phase, the matriarchal position of women was enhanced by their tradi-
tional association with agricultural magic or religion, women retaining
for a long time the character of priestesses. Nevertheless, this enhanced
matriarchal influence was unstable and comparatively brief, especially in
those regions where, as in Mediterranean Europe, the agricultural
revolution took place amid highly developed cultural contacts and
material industries. It was tradition rather than existing economic
conditions which favoured it. Women had long ceased to be the culti-
vators of the soil; their traditional ownership of it had been reduced to a
legal fiction. The assured food supply released the men from hunting.
The male yoked his oxen to the plough, and gradually took over the
bulk of agricultural labour. The household crafts which had hitherto
been almost exclusively in women’s hands, passed into those of men.
The sexual division of labour found in primitive societies was abolished.
Woman became economically unproductive and dependent. The con-
trast between the toiling primitive woman and the idle lady of civiliza-
tion, which has been mistaken for an indication of the enslavement of
the former and the freedom of the latter, indicates just the opposite. It
is the primitive toiler who is independent, and the unemployed woman
who has lost her freedom and is destitute.

One economic value alone was left to woman — her sex. In uncultured
societies little sexual competition exists among women, hence the
absence of ‘love’, i.e. individual preference; sexual selection is purely
economic. Primitive woman is therefore little disposed to cultivate the
arts of fascination — arts which have developed in proportion to the
decrease of her value as a worker. The woman who was no longer
economically self-supporting became competitive in terms of the only
value which remained to her, as an instrument of luxury and pleasure.
The Arabs forbid all manner of work to their daughters, lest their
beauty should suffer. The appearance of the females in a civilized

THE SOCIAL EVOLUTION OF MONOGAMIC MARRIAGE 209

society presents to the biologist a unique anomaly. While the male is
drab and inconspicuous, on the female’s attire all the resources of art
and wealth are lavished. To adorn her with the pigments and gloss of
secondary male sexual characters, birds and mammals the world over
are exterminated. The biological rule is reversed and so is the primitive
relation between the sexes — in the substitution of the patriarchal for
the matriarchal social order. Simultaneously the sexual instincts of the
man became discriminating. Regard for personal attraction, imaginative
desire, love, jealousy — all those sentiments which arc often regarded as
part of masculine sexual instincts — developed in consequence of the
revolution wrought by the plough. Women became the symbol of the
non-utilitarian values in which the fighter and the toiler sought refuge
from the harshness of reality.

The loss of woman’s economic value as a worker abolished the pur-
pose for which individual marriage originally arose. In those societies
which, after passing through a pastoral stage, reached a relatively high
stage of material culture, the increasing purchasing power of the men
enabled them to gather large harems of wives and female slaves; mar-
riage assumed a purely sexual aspect, and became the chief form of
sexual relation.

The position is different where, as in Europe, agriculture was not
preceded by a pastoral stage. Throughout a considerable portion of
European history the chief object of marriage, as in primitive societies,
remains economic, but in a somewhat different sense. It is not as a
worker that a wife is desired but as an heiress. The chief purpose of
marriage is to gain access to the property, to the lands, which in a matri-
archal social order were originally in the hands of women. Archaic
European marriage was thus governed essentially by economic rather
than sexual considerations. Hence, while polygyny was the Oriental
ideal, European marriage was of necessity monogamic.

Monogamic marriage is thus rooted in the special conditions which
led to European civilization. Religious and ethical conceptions became
grafted upon the economic institutions, and imparted to the latter a
religious and moral character.

In Europe the distinction between polygamy and monogamy has been
invested with transcendent importance. In the writings of the older
travellers scarcely any feature of non-European societies excites the same
zealous denunciations as does polygamy; every vice is traced to poly-
gamy, the head fount of all sexual immorality. This attitude is under-
standable, for the distinction goes to the root of the sexual code of
Christian Europe.

It was a firm belief in the sixteenth and seventeenth centuries that
polygamy had been invented by ‘Mahomet’, and its prevalence in Africa

210

THE MOTHERS

and in Asia was set down to the influence of Islam. The discovery that
polygamy existed in the New World was most unpalatable. Efforts were
made to suppress or minimise the fact; it was given out that representa-
tive North American peoples, such as the Iroquois and the Hurons,
were monogamous, or had been so ‘formerly’.

None of the European arguments against polygyny has any applica-
tion in the conditions of uncultured society. Early Christian moralists
who accounted marriage a necessary evil, favoured monogamy as a
reduction of that evil to a minimum, but chastity and continence are not
valued for their own sakes in uncultured societies. Monogamy accords
with the sentiments of exclusive attachment which are assumed in
European tradition to be the foundations of the union; but in primitive
societies such a desire is not understood. Polygamy is thought to imply
disregard for the feelings of women ; but in uncultured societies women
are the chief upholders of polygamy.

Absence of Jealousy amongst Women in Polygamous Families

Much evidence of the absence among primitives of the sexual jealousy
which moralists assume to be an inherent human trait might be ad-
duced, but a few examples will suffice. Among Kaffir wives ‘jealousy’,
says Delegorgue, ‘is unknown, and, far from dreaming of any sentiment
of the kind, the first wife of a Kaffir will work doubly hard, and to the
very limit of her strength, in order to acquire enough wealth to enable
her husband to buy another wife. This second wife, once acquired, is
bound to the first by bonds of affection for which we have no word in
our language. Those women are far more intimate than sisters’. Similar
denials of jealousy are made for the Ashanti and other African tribes, as
well as for New Guinea, Australia and the Pacific region, as also the
Eskimo and others.

In the accounts of travellers, we occasionally come upon references to
discord among the women in a polygynous family. But bickerings
between women who live and work together are liable to take place
whatever their relation to one another; and the occurrence of such
quarrels is as strong an argument against a multiplicity of servants or
shop assistants as against a multiplicity of wives. The disputes seem
mostly to do with economic matters. Women in West Africa, says Miss
Kingsley, do not care ‘a tinker’s curse’ about the relations of their
husbands with other women, provided that he does not waste on them
the cloth, etc., which they regard as their perquisite. If a man pays too
assiduous attentions to another woman, his wives insist on his marrying
her, so that she may share in the work of the household as well as in the
profits. In Australia a wife will sometimes show indignation if her
husband proposes to introduce into the household a woman older than

THE SOCIAL EVOLUTION OF MONOGAMIC MARRIAGE 211

herself, who might exercise the authority of age over her, but has no
objection to a younger woman. An Akikuyu woman, speaking to a lady
missionary, sent the following message to the women of Europe: ‘Tell
them two things’, she said; ‘one is that we never marry anyone we do
not want to, and the other is that we like our husband to have as many
wives as possible.’

Delager, writing of the Eskimo, mentions that he once asked a married
woman why her husband had taken another wife. ‘I asked him myself’,
she replied, ‘for I am tired of bearing children.’ The same thing has been
noted among the Koryak, the Chuckchi, the Omahas, the Kirghis
Tartars, the Ainu and among others. Hausa women encourage their
husbands to buy more wives because it adds to the respectability of the
family. Among the Ababua a wife ‘to prove the great affection which
she feels for her husband will, during his absence on a journey, buy him
a young and pretty girl whom she presents to him on his return’.

Polygyny is condemned chiefly on the grounds that it implies an outrage
on the feelings of women. Without doubt, were polygyny introduced in
a European country it would offend feminine feelings. But feminine
jealousy is a product of social conditions ; it is not a cause, but an effect.
It expresses the woman’s desire not to lose the male whom she wants as
an economic assistant and protector. It does not refer to the relations of
the male with other females, so long as they do not constitute a menace
to the economic association; and in a polygynous family, in primitive
and uncultured societies, they do not. On the contrary, since primitive
women are workers, it promotes the object of marriage.

No Condemnation of Polygamy outside European Countries

Biologically, polygyny provides the condition for efficient operation of
the male sexual function, the object of which is to impregnate as many
females as possible. Among bisexual plants one male serves for the
impregnation of numerous females. The male palm-tree stands sur-
rounded by an enormous harem of female plants. Among animals,
monogamous pairing is, as we have seen, extremely rare and occurs only
in special circumstances.

The European objection to polygamy is incomprehensible to uncul-
tured peoples and is interpreted as a sign of mental degeneration. In-
deed, polygyny is widely regarded as a moral virtue ; to support as many
fellow-creatures as possible is not only a mark of wealth but a form of
philanthropy. The number of a man’s wives is therefore the measure of
his respectability, not only in savage society but also in the highest non-
European cultures, as among Muslims and in China. The son of a
Sherbro chief, being asked how many wives his father had, displayed
considerable embarrassment, for his family was in somewhat decayed

212

THE MOTHERS

circumstances. He at last admitted that his father had ‘only twelve
wives’. Where economic conditions render polygamy difficult to achieve,
it remains, however, the object of ambition.

Uncultured races vary considerably in the extent to which polygyny
is practised, but no case is known, outside Christian nations, of a people
among whom it is morally reprobated. Before the Christian era, the
terms ‘monogamy’, ‘bigamy’, ‘polygamy’, in the sense in which we use
them, were unknown. The ‘prohibition’ of polygyny, which was
alleged to be ‘natural’ and ‘to be met with among all nations in a state of
refinement’, was actually promulgated for the first time in any part of
the world in the code of Justinian in the sixth century.

Limitations of Polygyny

Tribal custom is sometimes said to impose a limit on the number of
wives a man may have. Such statements are often couched in terms
similar to those in which the existence of polygamy is usually admitted
it is said to be ‘allowed’, ‘authorized’ or ‘tolerated’, but it is unlikely that
any uncivilized people ever thought to issue such an ‘authorization’.
To be sure, where women are difficult to obtain, a monopoly of wives
by any individual is sometimes resented. In some Australian tribes the
right of a man who is unable to procure a wife to challenge another who
has several is recognized. Among the Wasania of British East Africa, for
instance, men are not permitted to marry more than three wives, ‘as it is
considered that no man is able to provide food, etc., for more than this
number’. Elsewhere the figure is four (the number prescribed in Islam),
six or seven. Kings and chiefs, though privileged, are sometimes subject
to a limit. The number of wives of the king of Ashanti is said to have
been fixed at 3,333. Among the Jews, kings appear to have been limited
to forty-eight wives, although the number was commonly exceeded.
But neither economic necessity nor the natural objection of the majority
to the monopolizing of available women, appear to have led anywhere
to the enforcing of monogyny.

The term ‘polygyny’ is now employed in preference to ‘polygamy’ to
denote plurality of wives. The usage draws a useful distinction between
plurality of wives and plurality of husbands, or polyandry. But many
writers do not make use of the distinction consistently, for they do not
employ, in opposition to polygyny, the corresponding term ‘mono-
gyny’, but speak of limitation of marriage to one wife as ‘monogamy’,
thus conveying a misleading impression. The word ‘monogamy’ not
only denotes a practice, but also connotes a moral or legal standard
which precludes anyone having a multiplicity of wives. Statements that
a given people are ‘monogamous’, or ‘to a large extent monogamous’,
or even that (in a phrase of Westermarck’s) they are ‘generally rigorous

THE SOCIAL EVOLUTION OF MONOGAMIC MARRIAGE 213

monogamists’ (whatever that may mean), are liable to suggest that a
principle or law enforcing monogamy is recognized among those
peoples. But monogamy in this sense is not known outside Christian
nations. People among whom only one man in a hundred has more than
one wife are no more ‘monogamous’ than those among whom every
man has six wives.

Among primitives, to term a people polygynous or monogynous,
polygamous or monogamous, conveys little concerning the actual extent
and character of their sex relations. Even in Europe, it has been said,
monogyny has never existed. The Romans and the Greeks were mono-
gamous in their marriage institutions, but scarcely monogynous in their
sex relations. In Abyssinia, among the Christian Shoa, ‘monogamy, it is
true, is established by the Church, but concubinage is the habitual and
general custom, the king and his five hundred wives leading the way’.
In polygamous countries, where a man cannot afford to keep more than
one wife at a time, he makes up for the unfortunate circumstance by
frequent changes. ‘The Muslim of small means, who cannot afford to
marry two or more wives or to purchase slave-girls,’ observes Dr
Rohlf with special reference to Morocco, ‘compensates himself by
marrying one woman after another.’ In Egypt, partners were commonly
changed as often as twenty or thirty times in a couple of years. A
Bedawi will marry a young woman merely to enjoy her company for a
few weeks. A Baghdad dyer, who lived in strict monogamy, when he died
at eighty-five had had 900 wives ! Frequent changes of partners, where
polygamy is restricted, is equally conspicuous in the lower phases of
culture. Thus, the Guaycurus and other tribes of the southern Chaco
usually married only one woman at a time ; but the connections were so
transitory as to amount to general sexual promiscuity. As Dr Torres
observes : ‘ “successive polygyny” would be a more accurate designation
than “monogamy”.’ Many peoples, theoretically classifiable as mono-
gynous, because in old age they maintain monogynous marriage for
economic reasons, are actually totally promiscuous — for instance, the
Nagas, the Sakai forest tribes of Malaya, and the inhabitants of several
parts of Indonesia and Micronesia. Such ‘monogamy’, of course, affords
no sort of support to the view that sexual relations were originally
monogamous.

The number of wives which a man is able to support is limited by
economic conditions. ‘The poor man is a monogamist all the world over’,
says Weule. Frequently the number of children in a family is likewise
restricted by poverty. In several primitive societies a man is not considered
justified in rearing more than two or three children; the rest are killed.

The limitation of polygyny to the point of virtual monogyny is found
exclusively amongst forest dwellers, such as the Veddahs of Ceylon, the

THE MOTHERS

214

forest tribes of the Malay Peninsula, Borneo and South America, for
life is extremely difficult to support in the interior of forests. Such tribes
have no agriculture and few household industries; game is difficult to
pursue. In such conditions a number of women cannot contribute more
to a man’s needs and comforts than can one woman, and each additional
wife adds a burden to his resources. The position is radically different
among the Australian aborigines or the Fuegians, for instance.

The distribution of polygyny among uncultured races appears to be
governed by economic conditions exclusively. There is no reliable
evidence to show that the numerical proportion of the sexes plays any
important part in determining the extent of polygyny. (Estimates as to
that proportion among uncultured populations are extremely unreliable.)
Extensive polygyny is possible even where the number of females is
considerably smaller than that of the men, since in uncultured societies
females generally marry far earlier than males.

Estimates as to the extent of polygyny in a given community are
equally misleading. In any polygamous society most of the men will
probably be found at any one time to have only one wife, even though
all of them will later have two or more. A man who begins by having
one wife before acquiring three or four is no more a monogamist than an
unmarried man is a celibate. Hence the statements, so commonly made,
that ‘many men have only one wife’ or that ‘the majority of the men are
monogamists’ are quite misleading.

The expansion of Western civilization has enormously reduced poly-
gamy in every part of the world and in every phase of culture. Papuan
savages now have ‘an exaggerated idea of the wickedness of polygamy’;
and African negroes ‘call the ancestors of the tribe “polygamists” as if it
were a swear-word’. In India, among wealthy Hindus, in northern
China and Egypt, monogamy is becoming respectable, and in Turkey
has been made the rule. This is by no means due solely to the influence
of European opinion and of missionaries. The complete change in
economic conditions brought about by European expansion has been an
even more potent factor. The rise in the cost of living and of luxury has
rendered impossible the harems of mediaeval pashas. The higher
uncultured races are no longer dependent upon household industry, but
on European products paid for with the wages of male labour; and in
some parts of Africa those conditions have tended to abolish not only
polygamy but marriage. The most remote tribes, in the lowest phases of
culture, are no less affected. Their hunting-grounds have shrunk, their
life has become one of squalor. Among the American tribes of Canada
polygyny decreased within a few years of the first white settlements.
Nowhere do the conditions in which native tribes are found by the
traveller at the present day represent their natural state.

THE SOCIAL EVOLUTION OF MONOGAMIC MARRIAGE 215

Alleged Instances of Primitive Monogamous Institutions
The extent to which monogynous marriage occurs as a general usage
amongst uncultured peoples has been greatly exaggerated. Actually no
uncivilized people is known with certainty to have monogamous institu-
tions. In view of the prevalent misconceptions on the subject, we shall
examine a few of the statements on this topic, particularly those of Dr
Westermarck, which have been very frequently quoted without inquiry . 1

Thus, among the Guarani tribes, according to Dr Westermarck,
‘chiefs alone are allowed to have more than one wife’. The statement is
given on the authority of Father Charlevoix, who actually is specifically
referring to Christianized Indians. Westermarck adds a reference to
Father Hernandez. What Father Hernandez actually has to say con-
cerning the monogamy of the Guarani is as follows: ‘The Guarani
family, in their state of heathenism, suffered from a fundamental defect,
for polygyny reigned amongst them, and they thus violated the natural
law which is the basis of marriage.’ Father Ruiz de Montoya loudly
laments the unrestricted polygamy of the Guaranis; some of them had
as many as thirty wives. D’Orbigny says : ‘The customs of the Guaranis
are almost identical in all sections of the race. ... All of them practice
polygamy.’

Among some of the forest tribes of the upper reaches of the Amazon
basin, polygamy is reported to be rare or confined to the chiefs. Mr
Whiffen, to whom Dr Westermarck refers, and who in turn refers to
Dr Westermarck for theories of primitive monogamy, says that in some
tribelets south of the Tikie, chiefs have no more than one wife, but adds
the curious statement that ‘it is extremely hard to distinguish at first
between wives, concubines, and “attached wives”.’

Dr Westermarck mentions by name eleven South American tribes as
being, without qualification, ‘monogamous’; to wit, the Guaycurus,
Canellas, Shamboia, Paressi, Chavantes, Curetus, Purupurus, Mun-
drucus, Otomacos, Ackawoi and Macusi. But in eight of these cases
travellers have reported polygamous practices. For instance, the poly-
gamy of the Mundrucus was the chief obstacle met with by the Jesuit
missionaires ; their converts absolutely refused to give it up. Father
Ignace says that ‘the Mundrucus had for principle that it was licit to
have as many wives as their husband was capable of maintaining’. Bates
found the Mundrucus polygamous. We are therefore justified in regard-
ing with some reserve Wallace’s repetition of the same statement with

1 In the revised edition of his work. Dr Westermarck has modified his former
statements and eliminated several examples formerly adduced as evidence of
‘primitive monogamy.’ Thus, for example, the Iroquois were formerly
represented as ‘purely monogamous,’ and were repeatedly appealed to as a
conspicuous instance, statements now entirely withdrawn.

21 6

THE MOTHERS

reference to the Curctus and the Purupurus, of whom he says that he
never saw any and that ‘little is known of their domestic customs’. No
existing tribe is known by the name of Purupurus. According to Chand-
less, polygamy is practised by the chiefs among all the tribes of the Purus
river; among the Ipurinas it is general. Von Spix and Martius did not
form a high opinion of the sexual customs of the Curetus, who pressed
their daughters on the travellers.

Father Gumilla is the source for the assertion that the Otomacs are
wholly monogamous, but Baron Humboldt, who obtained full informa-
tion from the missionaries living amongst them, speaks very dis-
paragingly of Father Gumilla’s account and says that ‘all the Indians
who will not be baptized live in a state of polygamy’. Father Gilii con-
firms this. Azara saw one of the last representatives of the Guaycurus :
he was living with three wives.

In no instance is the alleged monogamy of any of those South
American tribes established with anything like certainty. It is striking
that an authenticated instance of a monogynous tribe should be so
difficult to find where the aboriginal population has been broken up into
innumerable warring tribelets and on the verge of starvation and extinc-
tion, conditions in which we might well expect to find monogyny.
Many of the poorer tribelets consist of scarcely more than three families.
When, in such tribelets polygyny is said to be confined to the headman,
this represents a greater proportion of polygyny than is to be found in
any Muhammedan country.

Again the Guanches of the Canary Islands are very generally described
in modem accounts (including Westermarck’s) as having been mono-
gamous. The source of those statements is probably the work of Father
Abreu de Galindo, who wrote towards the middle of the seventeenth
century, when any surviving remnants of the Guanche race were
Christians. His account is extremely edifying, but contradicts all the
older accounts. Friar de Espinosa, who dwelt many years in the Gu-
anche community of Candelaria, states that the Guanches ‘had as many
wives as they pleased and could support’. Gomara, Sir Edmund Scory,
Galvano, Cadamosto and Bemaldez all confirm this.

The monogamy ascribed to some Berber tribes meagrely offsets the
polygamy prevalent throughout Africa. Thus the Tuareg have several
times been stated to be ‘strictly monogamous’. But, if correct, the state-
ment can apply to some of the north-western tribes only, for in the
southern and eastern Sahara, the Tuareg ‘usually have from two to four
wives’. The monogamy of the northern Tuareg has been supposed to be
a relic of their former Christianity. If so, it was certainly an innovation.
‘Among the Numidians and the Moors,’ Sallust tells us, ‘each man
marries according to his means as many wives as he can; some two,

THE SOCIAL EVOLUTION OF MONOGAMIC MARRIAGE 217

others more, and the kings many more.’ The nomadic Tuareg commonly
have establishments in the various villages which they visit; hence the
appearance of monogamous households which has so often deceived
travellers ; in any case, they keep concubines in addition. Many do not
marry at all, and are satisfied with a harem of purchased girls.

The western Sahara, or Moroccan region is, with the exception of a
few patches, a land of extreme poverty. The men are haggard with
hunger, and whole populations are decimated by famine. In such condi-
tions it would be surprising if large households were common. Never-
theless, there is little definite evidence of general monogyny, except for
a few communities, such as the Dads of the lower Atlas, and polygamy
is found in every district. Dr Westermarck cites Chavannc, who refers
to Vincent as stating that he ‘did not meet a single man who had a
plurality of wives’. Dr Rohlfs met one at Tafilet, in the heart of the
same region, who had three hundred; and Mr W. B. Harris, who per-
haps knows that region better than any other Englishman, speaks of the
harems and of the large polygamous households and slave-girls of the
Sharifian families.

From India, Dr. Westermarck has not succeeded in culling a dozen
instances of tribes alleged to be monogynous. Among these are the
Khasis. ‘The practice of polygamy’, says Mr Gait, ‘is usually said to be
uncommon amongst them’; but he adds: ‘an educated Khasi whom I
consulted assures me that polygamy is by no means unknown. It was
formerly considered meritorious for a Khasi to beget offspring by differ-
ent wives.’ Among the Nagas, who are also alleged to be monogamous,
‘polygamy is very common, and is limited only by the men’s resources’.
The Mikirs, who are cited by Dr Westermarck as monogamous, are
expressly stated by Mr Stack to be polygamous. It is noteworthy that
polygamy in India is characteristic of the more primitive rather than of
the more advanced races. In southern India, the most ancient aboriginal
inhabitants are supposed to be the Yenadie, Villee and Vede; they are
polygamous.

‘The early discoverers of the Philippines’, says Dr Westermarck,
‘found legal monogamy combined with concubinage.’ What the early
discoverers of the Philippines found, according to one of the oldest
accounts, was that the natives ‘marry as many wives as they can afford to
keep’; what Magellan found was that ‘they have as many wives as they
wish’; what de Legazpi found was that ‘the men are permitted to have
two or three wives if they have money enough to buy and support
them’. What the discoverers found in the Bisayan, or Middle Islands,
was that ‘all the men are accustomed to have as many wives as they can
support. The women are extremely lewd, and they even encourage their
own daughters to live a life of unchastity’. Father Chirino assures us

218

THE MOTHERS

that ‘we are gradually uprooting that hindrance to conversion so com-
mon among those people and so difficult to remove, the practice of
having several wives’. Dr Westermarck’s remarks concerning the
Subanu, the Italons, the Tinguianes, the Tagbanuas of Palawan and the
Negritos of the Philippines are equally ill-supported.

The Igorots of Luzon have been specially instanced as an example of
‘primitive monogamy’, and also of a ‘lofty ideal of chastity’. Various
authorities contradict this, and the missionaries expressly denounced
the polygynous habits of the Igorots. ‘In case of adultery’, says Dr
Westcrmarck, ‘the guilty party can be compelled to leave the hut and
the family for ever.’ Dr Jenks says that there is no tribal law against
adultery, and that married men commonly frequent the ‘olags’ of the
unmarried girls who solicit them.

We have seen that the ‘primitive chastity’ which has been ascribed to
the forest tribes of the interior of the Malay Peninsula is as imaginary as
that alleged of the Igorots. Among the Mantras (a tribe of the Jakuns)
polygyny. Dr Westermarck tells us, ‘is said to be forbidden’. ‘It is
nothing rare’, says Father Bourien, ‘to meet individuals who have been
married fifty times’; and, in spite of the alleged prohibition, some
nevertheless live in simultaneous polygamy. ‘Most of the Binua,
according to Logan’, says Dr. Westermarck, ‘have only one wife’, but
what Logan actually stated was : ‘Most of the Binuas have only one wife,
but some have two, and there does not appear to be any rule on the sub-
ject’; and he adds that separation is easy, that husbands commonly ex-
change wives, and that adultery is frequent and unresented.

‘The Central African pygmies’, says Dr Westermarck, ‘seem to be
mostly monogamous . . Mgr Le Roy, who is perhaps one of the most
competent authorities on the Central African Pygmies, says that
‘polygamy is not condemned among them; on the contrary, if it is not
universal, it is because it is not always possible’. Sir Harry Johnston
says of the pygmies of the Tanganyika region: ‘According to the evi-
dence which I have myself collected, they seem to approach very near to
promiscuity, and even incest, in their marital relations.’ Similarly, the
Bushmen have as many wives as they can afford, despite Dr Theal’s
assertion that they are monogamous.

Dr Westermarck has not been able to discover any suggestion of
monogamy in Polynesia or in Melanesia. The Rev. W. Y. Turner
received the impression that the natives of Motu were monogamous,
but subsequently discovered that, in addition to the wives with whom
they happened to be living, the men had other wives in the next
village.

Nowhere among the Australian aborigines is there even a tendency
towards monogyny. ‘Polygamy to the fullest extent is an Australian

THE SOCIAL EVOLUTION OF MONOGAMIC MARRIAGE 219

institution 5 . Dr. Wcstermarck says that Mr Curr ‘has discovered some
truly monogamous tribes’. But nobody else has, and Dr Malinowski
contradicts him, saying ‘polygyny seems to be found in all the
tribes’.

The customs of the now extinct Tasmanians appear to have been
practically identical with those of the Australian aborigines. ‘Plurality
of wives was the universal law among them’, says Mr Lloyd.

It would be difficult for any hypothesis to be so uniformly and directly
in contradiction with the facts than the theological doctrine of primitive
monogamy. ‘The evidence’, say Messrs Hobhouse, Wheeler and Gins-
berg, summing up their collection of reports, ‘does not make for the
association of monogamy with the lowest culture, but only of monogyny
with one particular form of that culture (the forest tribes), and that only
partially’. Their extensive tables, though containing several disputable
entries and open to certain criticisms, exhibit the general distribution of
polygyny very well. The factors which are supposed to make for mono-
gamy, the authors observe, ‘in every grade, when considered as a whole,
are seen to be overborne by the opposite forces making for polygamy’.
The number of a man’s wives reach as a maximum in pastoral societies,
which are invariably polygamous.

While polygamy is everywhere disappearing with the advent of more
complex economic conditions, there is no reliable instance of a people
among whom monogamy was once the rule having adopted polygamy.
Most allegations that a given people were ‘formerly’ monogamous, such as
that of Father Charlevoix concerning the Hurons, or of Major Cremony’s
concerning the Apaches, are manifestly destitute of significance. When-
ever it is stated on more solid grounds that a change from monogamous
to polygynous marriage customs has taken place among a given people
it will be found, I think invariably, that their earlier condition was one
of unrestricted relations within the permissible degrees.

Thus, Wcstermarck mentions the Alfurs of the Minahassa peninsula
of North Celebes as having formerly been monogamous ; and he might
have mentioned a much more authoritative statement of Professor
Wilken to the same effect concerning the Alfurs of Buru. But the coastal
tribes of the Alfurs are certainly not their primitive state. There are,
however, several examples of tribes of the same race which have re-
mained isolated from external influences. Their ‘monogamy’ bears little
resemblance to what we understand by monogamous relations. Thus
Heer Sachs admits with reluctance that he was unable to discover among
the Alfurs of the interior district of Setie, in Ceram, anything that can be
properly termed marriage. ‘They live in a state of free love. The woman
keeps company with the man of her choice for such length of time as it
pleases both. 5

220

THE MOTHERS

Status of Different Wives in Polygynous Families

The women in a polygynous family are often spoken of in reports as
‘concubines’, with the exception of the one first married. Actually, the
distinction between wives and concubines belongs in general to ad-
vanced stages of society, although some war-like tribes draw a distinc-
tion between captured female slaves and wives of their own nation. In
some uncultured societies one wife (not invariably the one first married),
is referred to as the ‘chief wife’. It has been suggested that in such cases
the ‘chief wife’ is in reality the only ‘real’ wife, and that the others are
little better than ‘concubines’, thus implying that such marriages,
although in appearance polygamous, are theoretically monogamous.
But it can be shown that development has commonly taken place in the
opposite direction. Where the wives are sisters, the eldest sister natur-
ally exercises over the younger ones a certain authority and enjoys a
certain precedence, which would be the same whether they were mar-
ried or not; and an older woman usually has the privileges of seniority.
But these distinctions are different from that between the ‘chief wife’
and other wives which is found in some societies, and which observers
are prone to detect in uncultured peoples, among whom it is very rare.

Reliable reports from all over the world stress the complete equality
of wives. The wife who is spoken of as the ‘chief wife’ is very frequently
merely the favourite for the time being, and very often the last married.
Thus among the Sioux, the oldest wife, being generally the eldest
sister, enjoyed a certain precedence; but the one who was last married
had an equal status. Among the Chukchi, and the Eleut Tartars an old
wife is deposed and becomes the servant of the younger ones, or is
driven from the home.

Distinctions between the various wives are found chiefly in Africa
and Asia. Nevertheless, wives are on the same footing among the
Central African pygmies, the Congo tribes, in Angola and Abyssinia,
among the Banaka and Bafuku of the Cameroons, etc.

In Africa the term ‘chief wife’ commonly refers to the distribution of
agricultural duties among the wives. A ‘chief wife’ is needed to direct
the work of the others, like a foreman. In Senegambia, if the ‘chief wife’
proves incompetent, her position is given to another wife. Among the
tribes of British Central Africa, a man may change his ‘chief wife’
whenever he thinks fit. Frequently, when a man possesses extensive
fields, he has to appoint several ‘chief wives’. Among the Medge, when
a man’s wives become too numerous to live in one hut, he divides his
establishment, appointing a ‘chief wife’ for each group. Such practices
are scarcely indicative of monogyny.

In certain advanced social conditions, however, the distinctions

THE SOCIAL EVOLUTION OF MONOGAMIC MARRIAGE 221

between a ‘chief wife’ and the other wives may have a real juridic sig-
nificance; and one which has an important bearing upon the develop-
ment of monogamic institutions. A woman of higher social rank natur-
ally enjoys a superior status; such status is independent of whether she
has been married before or after other wives and of the husband’s
predilections. Among the Binjhalsa of Bengal a man is bound to take a
new wife when he succeeds to the rank of ‘zemindar’, or landowner,
even though he is already married ; the new wife is the ‘pat rami’, or
‘principal wife’. The social rank of a man’s first wife, who is usually
selected for him by his relatives, is regarded as of the highest impor-
tance. It is obvious that the trouble and expense of securing a wife of
suitable social qualifications can rarely be undertaken more than once.
Hence it is a general rule that such social and mercenary considerations
are taken into account in regard to a man’s first wife only. These require-
ments once satisfied, a man must, as a rule, be content to select his sub-
sequent wives by such less reputable considerations as love, youth,
beauty and industry. The position of a man’s first wife is thus often,
quite apart from any relative status bestowed upon her by her marriage,
socially superior to that of his subsequent acquisitions.

These considerations acquire an enormously enhanced importance,
when, as in many advanced societies and in early patriarchal societies,
the paramount consideration comes to be the breeding of legitimate
heirs. A ‘legitimate’ wife is primarily a wife who can give birth to a
‘legitimate’ heir. It is the legitimacy of the heir which makes the wife
‘legitimate’; a wife who is barren is not regarded as legitimate, and can
in all lower cultures be divorced. Sometimes — as, for instance, among
the Maori — when no distinction of social rank exists between the wives,
the ‘chief wife’ is she who first gives birth to an heir. Where primitive
tribal organization is still strong but is passing from a matriarchal to a
patriarchal order, a ‘legitimate’ child means a child which belongs to his
father’s and not to his mother’s clan; the breeding of illegitimate chil-
dren in this sense defeats the main object of procreation, the increase of
the tribe and its perpetuation. The breeding of such ‘illegitimate’
children is consequently as great an offence against patriarchal society as
the transference of women to another tribe is against matriarchal society.

Islam condemned the once-common free contract between a man and
woman, known as ‘al-motah’, because it gave the man no offspring that
could be reckoned to his tribe and not to the tribe of the mother.

One of the merits of polygyny in the eyes of uncultured peoples is that
it increases the number of a man’s progeny. But where the transmission
of private property becomes an important consideration, it is incon-
venient to have numerous heirs, and the bulk of the property, or the
whole, is sometimes transmitted to the children of the ‘chief wife’ only.

222

THE MOTHERS

In such conditions — and they are the conditions of settled agricultural
society — what is desirable is not a large progeny but an heir, and what is
required is not a large number of legitimate wives but one legitimate
wife. That other wives should be ‘legitimate’ is not only of no impor-
tance but is positively undesirable.

Obviously, the distinction in rank between the mother of the legiti-
mate heir and other wives must become more sharply pronounced as
private property grows more important. This differentiation will be
accentuated by the reluctance of self-respecting families to allow their
daughters to occupy an ever more disreputable and ‘illegitimate’ posi-
tion.

In settled civilized society, the wife thus acquires an entirely new
function. Whereas among primitive people she is primarily an economic
associate, a provider of food, a labourer, and thus the chief source of
wealth, in pastoral and higher agricultural civilization, only her sexual
value is left. This sexual value consists in her attractiveness, in her
function as an instrument of pleasure; the beauty of idle women is
cultivated, and they are gathered together in large harems. Settled
agricultural civilization bestows upon woman a new function and a new
sexual value — that of legitimate wife, of mother of legitimate heirs to
property.

The contrast which we have noted between pastoral and purely
agricultural societies is thus accentuated. Pastoral property is readily
subdivided and multiplies ; the importance of transmitting undivided to
one or a few heirs is not pressing. Essentially pastoral peoples, such as
the Semites and the Arias of Asia, have in general retained their pastoral
traditions with extreme conservatism. Yet, even here the acquisition of
other forms of property has tended to reduce the extent of polygyny. In
India, in like manner, polygamy has tended to become restricted.
Modem economic conditions, together with the pressure of European
opinion, are converting the polygamous institutions of the East into
virtual monogamy.

Polygamy and its Decay amongst the Jews

An evolution similar to that now taking place in the Orient took place
under the Roman Empire among the Jews. Like all other Semitic
peoples, the Jews were polygamous; the harems of Jewish kings were
vaunted for their exuberance (Genesis iv. 19; Judges, viii. 30; etc.).
Nowhere in the fastidious Jewish law is there any condemnation of
polygamy. Hebrew Law assumed polygamy. Among later Talmudist
doctors opinion varies : some recommended that a man should not have
more than four wives ; others thought he might have as many as he could
afford. Moses Maimonides said: Tt is lawful for a man to marry as

THE SOCIAL EVOLUTION OF MONOGAMIC MARRIAGE 223

many wives as he pleases, even to a hundred, either simultaneously or
successively/ In Italy as late as the sixteenth century a Jew who had
had no children by his wife married another, a formal dispensation for
the bigamy being obtained from the Catholic Church. In Spain, Jews
were polygamous as late as the fourteenth century, and still are so in
some polygamous countries. The first pronouncement against polygamy
among the Jews was made in the eleventh century at Worms, but was a
merely local resolution. Since no law has ever existed against polygamy
among the Jews, should a Jew in England at the present time contract a
bigamous marriage, it could not be dissolved according to Jewish law
without a formal divorce.

The object of Jewish marriage, as of all marriage in a settled order,
was to obtain a male heir; hence, as in lower cultures, a barren wife
might be divorced, or she might evade this by supplying another wife —
for instance, one of her slave-girls. The offspring, of course, was
accounted the son and heir of the husband. A male child was accounted
the son of the chief wife, no matter who was the actual mother. Hence,
respectable families became unwilling to sell their daughters except as
‘legal’ wives.

The Oriental Harem

The undignified status of secondary wives depends upon the accentua-
tion of the distinction between them and the principal wife, and not on
the fact of polygyny. In the most highly developed juridical polygamous
families, such as those of the Muslim world, where the distinction of
wives has no juridic significance, the position of women is very far from
being one of degradation. Popular Western notions concerning the
Oriental harem are, of course, little more than a tissue of myths.

Competent persons agree that the position of women in Turkey is,
and was, actually higher, freer and juridically superior to that of women
in Western Europe. T think I never saw a country’, said Lady Craven
over a hundred years ago, ‘where women enjoy so much liberty and are
freer from all reproach as in Turkey.’ The Turkish woman is free to go
and come as she pleases; she spends most of the day out of doors,
usually absents herself for the night when visiting some friend, and is
never called to account for the way she spends her time. From a juridic
point of view, the position of Turkish women is certainly much higher
than that of English women before the passing of the Married Women’s
Property Act . 1 A Turkish woman has equal rights of inheritance with
her brothers. She has the uncontrolled disposal of her own wealth and
property, and can inherit property without the intervention of trustees.

1 Briffault is, of course, speaking of Turkey before the reforms of Ataturk.
G.R.T.

THE MOTHERS

224

She can sue, or be sued, independently of her husband, and also sue,
or be sued, by him. Although freedom of divorce is ostensibly greater
on the part of the husband, yet the wife can obtain it without difficulty.

The Oriental woman enjoys, as a rule, more consideration and more
deference from her husband than the average European wife. The great
difference lies, of course, in the complete segregation of the sexes and
consequent absence of intersexual social intercourse. This is not an Arab
or an Islamic institution, but was introduced into Islam from Persia,
where it had been customary from remotest antiquity. The Persian
harem — the word means ‘sanctuary’ — derives not from any degraded
status of women, but from the almost superstitious reverence with
which they were regarded. All records bear witness to the high position
of women in ancient Persia, a position which amounted to gynaxocratic
domination. A man might not even sit down in the presence of his
mother without her permission. The original harem was in all prob-
ability the matriarchal home of the mother, with whom her children
remained, the father being excluded from even seeing them until they
were five years old. ‘The harem’, says von Hammer, ‘is prohibited to
strangers not because women are considered unworthy of confidence,
but on account of the sacredness with which customs and manners
invest them.’

Much of that predominant position persists in Persia at the present
day. As in Turkey, there is in fact very little real seclusion of women in
Persia. Every woman goes as a matter of religious duty, each day to the
‘hammam’ or bath ; and this is a sort of club where she meets her friends
and usually spends seven or eight hours. The position of women in
Persia is indicated by the form of the marriage ceremony: the woman
places her right hand over that of her husband ‘to show that she ought
always to have the upper hand of her spouse’.

Inevitably, the harem has at times degenerated into an instrument of
male domination ; and seclusion has undoubtedly narrowed the mental
outlook of the Arab women, formerly conspicuous for her achievements.
But this atrophy has been greatly exaggerated. The frivolity and mental
undevelopment of the harem woman is not in general greater than with
many women of fashion in Europe. While the aristocratic lady cultivates
such arts as embroidery, many of the Persian middle-class women are
highly educated according to Oriental ideas. They read, and often write
poetry; they sing, play and do plain and fancy needlework. The wife is
consulted in all things to a much greater extent than is the case in Europe.

In the Oriental polygynous family, although there is a ‘chief wife’,
there is no real social or legal distinction of rank between the several
wives. Indeed, it is a current Islamic doctrine that the True Believer
who displays any partiality for one wife above the others will incur a

THE SOCIAL EVOLUTION OF MONOGAMIC MARRIAGE 225

special punishment at the Day of Judgment. There is no juridic distinc-
tion whatever between the children; every woman who bears a child
becomes thereby the legal wife of the father. There are consequently no
illegitimate children in T urkey. If the man has already four wives, the
woman becomes an odalisque, but her children have the same rank and
rights as the children of the ‘chief wife’. If the woman is a slave,
motherhood makes her a free woman.

The evolution from polygamy to monogamy has been completed
among the Osmanli Turks in our own day — not by reduction in the
number of ‘legal’ wives, but by the gradual elimination of concubines.

Marriage and the Position of Women in China

In China, in contrast, the differentiation between the chief wife and
other wives became so complete as to constitute juridic monogamy,
while the Chinese woman gradually sank in status to perhaps the most
depressed level found in any cultured society.

Chinese marriage institutions are virtually identical with those of the
Mongols and Turkis of Central Asia. The same elaborate ceremonials
involving ‘go-betweens’ arc found in both ; and the fact that these cus-
toms are regarded as immemorial institutions among the remotest
tribes precludes the supposition that they were adopted from the
Chinese. Among all Mongols, the same verbal distinction is made
between the ‘great’ wife, or ‘tsi’, the ‘little wives’, or ‘tsie’, and ‘concu-
bines’, and has practically the same significance as in China. The
Mongols are frankly polygamous ; yet such is the emphasis laid upon the
position of the ‘great wife’ that they regard themselves as monogamous.
The children of the ‘little wives’ have no legal right to a share in the
inheritance, though they may be given one. Among the ruder Western
Mongols we find the same situation; but, according to an early eigh-
teenth-century account, the significance of the distinction between
‘great wife’ and ‘little wives’ and concubines was then very much looser.
‘It is a constant rule among the Tartars, who look for nothing but youth
in their wives, to give over lying with them when they draw near forty
years, reckoning them no more than old housewives, to whom they give
their victuals for taking care of the house and tending the young wives
who may occupy their place in their master’s bed.’ Thus, the hierarchical
distinction, so important in China, can scarcely be said to exist; the
‘great wife’ is merely, as in many African families, the favourite wife for
the time being. Among the Kirghis the conditions are much the same.
Among the Mongol Khans the legal distinctions up to the twelfth cen-
tury were exactly similar to those noted among the Eleuts and the Kir-
ghis. But no indignity whatever attached to the status of ‘little wife’.
Powerful khans gave their daughters in marriage to other khans irre-
t.m. — 8

226

THE MOTHERS

spectivc of whether they went as ‘great wives’ or as ‘little wives’.
Jinghis Khan had over 500 wives and concubines, and all his wives were
daughters of khans. His first wife was termed the ‘mother of his sons’;
his second wife was the daughter of the Emperor of the Manchus. Even
children by concubines were regarded as perfectly legitimate, those of
the legitimate wives merely taking precedence over them in the order
of succession — though they could also inherit.

The usages of the ancient Chinese chieftains were identical with those
of other Tartar khans, and it was common practice to marry two sisters
or two cousins. Polygamy is the law of the land and is by no means
confined to the well-to-do. Marriages of ‘little wives’ are performed
with the same pomp and solemnity as with the ‘great wife’; and they are
officially presented to friends and relatives. The distinction between
‘great’ and ‘little’ wives is a purely juridic one, intended to preserve the
unity of the inheritance. Hence ‘little wives’ cannot inherit, although
their children, who bear their father’s name, do so under the legal fiction
that they are the children of the ‘great wife’, married with the assent of
the ‘great wife’, and frequently at her request.

Marriage becomes indissoluble after the first three years. The union
is contracted not only for this world but for the next. Formerly the wives
of the emperor were killed at his funeral, as in Africa. Until lately a
widow was expected to commit suicide on the death of her husband. It
was, and is still, the correct thing for the announcement of her suicide
to appear in the papers beside the death-notice. A widow may not marry
again, nor may she return to her own family.

The position of women in China was, as we saw, formerly much
higher and freer. Among Mongol tribes, both eastern and western, it is
one of considerable independence and influence. ‘In the household’,
says Prejevalsky, ‘the rights of the wife are nearly equal to those of the
husband’; and the latter is sometimes positively henpecked. Women
have the right of divorce. Tartar ladies are viragoes, who may sometimes
be seen pursuing a terrified husband with a whip. Throughout Tartar
society the mother occupies a position hedged about with reverence and
awe. Even the redoubtable Jinghis Khan, the ‘scourge of Asia’, ‘the
conqueror of the Universe’, quailed before his mother and was compelled
by her to alter his policy.

In historical and modem China, on the other hand, the status of
women is abject; according to the Confucian dictum: ‘The woman
obeys the man: in her youth she obeys her father and her elder brother;
when she marries she obeys her husband; when her husband is dead she
obeys her son.’ The education of a girl, according to the rules of the ‘Li
Ki’, consists in learning to spin and to be submissive, docile and
obedient; a boy is taught to speak boldly and clearly, a girl submissively

THE SOCIAL EVOLUTION OF MONOGAMIC MARRIAGE 227

and low. A woman may be divorced because she is too talkative.
‘Ignorance and retirement are proper to a woman/ Most Chinese
women are illiterate. In the olden days, exceptions were common in
mandarin families. The most famous is Pan-Hoi-Pan, who wrote a
celebrated treatise on the duties of women. ‘We occupy’, she says, ‘the
last place in the human species, we are the weaker part of humanity; the
basest functions are, and should be, our portion.’ She proceeds in the
same strain through seven learned chapters, citing authorities. From the
age of seven the seclusion of Chinese women is absolute. Up to the
seventh century women were veiled. Married women go out only in a
sedan-chair with the blinds drawn — though women of the highest rank,
when attended by a numerous train, have the blinds drawn up. At the
present day [1927] if a woman or a girl should be seen in the open street,
she is subjected to obscene remarks. In the house, it would be the
grossest insult for a visitor to inquire after his host’s wife. A woman may
not eat with her husband or with her son, and she may not even remove
the dishes from which they have eaten, for her touch would defile them.
When a man entertains friends in his house and female company is
desired, courtesans are called in.

It is not surprising that societies have frequently been formed of girls
who are sworn to celibacy, and who vow to kill themselves rather than
accept marriage except on purely matrilocal terms. ‘It is not an un-
common thing’, says Mr Dyer Ball, ‘to read of girls in various parts of
China committing suicide rather than be forced into marriage.’

Chinese marriage has nothing whatever to do with affections or
sexual relations. It is regarded as a duty for the perpetuation of the
family and the transmission of property. It is usually entered into as
late as possible.

In Korea, marriage customs are identical with those of China. The
marriage relationship is briefly summarized in the remark of a Korean
gentleman: ‘We marry our wives, but we love our concubines.’

As a consequence, the professional courtesan has attained the highest
degree of development. There are various different classes of ‘daughters
of flowers’ as they are called. ‘One must not confound the accomplished
and learned Chinese courtesans’, says M. Bazin, ‘with those who
“publicly display their smiles”, as the Chinese poets have it, and run
after pleasure.’ She must be proficient notably in music, dancing, history
and philosophy, but must also be able to write the tao-te-king characters
of the book which contains the doctrine of the philosopher Lao-tse.
When thus qualified, she becomes a “free woman”. She is emancipated
,/rom the duties of her sex, and may well account herself superior to the
legal concubine and the legitimate wife. Many Chinamen of the better
class marry such women, who become devoted wives.

228

THE MOTHERS

Greek Marriage

The evolution of marriage in China is of interest because of its striking
similarity to conditions in ancient Greece. The early Greeks were polygy-
nous. In Homer and the tragic poets, the Trojans are represented as having
legal wives, the Greeks as having ‘concubines’. Polygamy was the rule in
Troy. Priam had many honoured wives; the progeny of Antenor, wisest
of the Trojans, was brought up by his chief wife, Theano, with her own
children; Andromache even suckled the children of Hector’s other wives.

The households of the Greek chieftains are described as abounding in
‘concubines’, but no great weight can be attached to the distinction:
Thesaeus is expressly mentioned as having married his several wives ‘in
legal marriage’. Even in historical times Dionysius of Syracuse had two
legal wives and Philip of Macedon had numerous legal wives, many of
whom were princesses.

There was a very good reason why, in archaic Greece, secondary
wives should have fallen to a subordinate status, and why one wife only
should have come to be regarded as ‘legitimate’. It was the princess
whom a chieftain married who bestowed upon him rights to the throne
and to her possessions ; even Agamemnon only held his throne through
his wife’s rights. It would, obviously, have been impossible to have
more than one such wife. The wife proper did not resent the presence of
concubines provided that they were of lower rank.

In historical Greece, polygamy was never prohibited, and there is no
reference in Greek literature to its being regarded as immoral; it was
merely looked upon as a non- Hellenic custom. Indeed, during the
Peloponnesian war, when Athens was depopulated by plague and
casualties, bigamy was enjoined by law. The Greeks were not at all
eager to marry. In his poem, ‘Works and Days’, Hesiod offers much
Polonius-like advice to his younger brother, Perses, and counsels him to
settle down to the quiet rustic life of a small farmer : he should procure a
house and allotment, a ploughing-ox, and also ‘a woman, purchased, not
wedded’. Above all, he warns him against the enticements of a woman
‘with waggling rump’ that should ‘seek after his home’. Marry some day
he certainly should, but not until he has reached a mature age, at least
thirty. It is also advisable that he should marry a virgin, ‘so that you may
teach her chaste morals’.

In Greece, as in China, marriage was a purely juridic procedure of
which the object was that ‘the heritage should not be left desolate and
the name cut off’. In lieu of the right to the chieftainship and lands which
heiresses bestowed in archaic times, the Athenians offered a dowry as an
inducement. Medea, in Euripides, complains that ‘We have to buy our-
selves husbands at great cost.’

THE SOCIAL EVOLUTION OF MONOGAMIC MARRIAGE 229

Between the dowry and the bride-price there is a developmental
difference. When property accumulates in the hands of men, they will
use it to ‘purchase’ wives ; when it remains in the hands of women, they
will ‘purchase’ husbands. The former, as we have seen, occurs in
pastoral societies, such as the Arabs, the Jews, the Indian Aryans and the
Tartars; the latter where arable land is the chief form of property, and
the society is still matriarchal. This is precisely what we find in regard
to the transmission of princely inheritance in archaic Greece. It was
transmitted in the female line ; and the sons of princes went forth from
their home to marry princesses in order to obtain princely rights.
Accordingly, the ‘dowry’ remained throughout the social history of
Greece the pivot of legal marriage.

In historical times the position of the Greek wife was identical with
that of the Chinese wife. She received no education beyond being
taught by her mother to spin, to weave and to cook. She was ‘not allowed
to see, hear or ask anything more than was absolutely necessary’. She
should not appear when her husband had invited guests, nor should she
appear in public. Theban women were forbidden to walk in the street.
No woman could inherit property; an ‘heiress’ did not herself inherit
property, but had, on the contrary, to be inherited; the property lay
fallow so long as the woman had no guardian or owner. When a man
died his property might go to the most distant male relative but never
to his wife, who had to return to her own people. In fact, a woman
never really became part of her husband’s family; Greek marriage
remained juridically matrilocal; even after marriage, a woman remained
under the guardianship of her father. On the other hand, a man might
transfer his wife to a friend.

Greek marriage had no connection with love. Ottfried Muller
knew of no instance of the marriage of a free-born Greek woman
for love. Indeed, marriage and love were regarded as quite opposed. In
a play of Terence (a translation from the Greek) the lovers lament
that they are going to be married: ‘Every lover feels it to be a sad
grievance that a wife should be assigned to him.’ Aristotle, endeav-
ouring to put marriage in the most favourable light, speaks of friend-
ship growing up between husband and wife ; neither he nor any other
author mentions love. The wife in Greece was a housekeeper and
the bearer of lawful heirs ; she was not the sexual companion of the
man.

For his sexual life the Athenian Greek relied on ‘hetairai’. The
character and status of those women has been the subject of contra-
dictory accounts, some representing them as being cultured Aspasias,
others as common prostitutes. The truth is that there were all sorts. The
word hetairai, which means ‘companion’, was, by a euphemism, applied

THE MOTHERS

23O

to all . 1 Originally it was a perfectly honourable appellation. Doubtless
its use dated from a time when the two orders of union* the patriarchal
and propertied ‘legal* marriage and the free matriarchal union by
mutual consent* co-existed. This organization persisted in point of
fact throughout the hey-day of Greek civilization. No social stigma
attached in Greece to sexual freedom* because the standards of sexual
morality had not yet developed. Matrons themselves had no scruple in
associating with hetairai. They were honoured, as any other persons
might have been, for distinction of talent, or of conduct* or of beauty.
No obstacle lay in the way of their social advancement. Perikles sent
away his 'legal’ wife and installed Aspasia in his house; Myrrhina
shared with Demetrios, king of Macedon* all but the crown; Thargelia
married Antiochos* king of Thessaly; Pitho married Hieronymos and
became queen of Syracuse. Thais was the companion of Alexander, and,
after his death, married Ptolemy* the first Lagide king of Egypt.

The two opposite types into which woman, emancipated from eco-
nomic production* had developed* are thus seen in Greece, clearly
contrasted. Her two economic values, her two functions, as sexual
companion and as mother in relation to the transmission of legal rights
and property* remained separate. The attributes of each function were
totally differentiated. The hetaira* elegantly adorned and highly edu-
cated, was a being of a different race from the cloistered housekeeper
and breeder of heirs, to whom adornment was forbidden and whose
mind was stunted. 'We have hetairai for our delight,* says Demosthenes,
‘concubines for the daily needs of our bodies, wives in order that we may
beget legitimate children and have faithful housekeepers.’

Roman Marriage

Roman marriage differed considerably from Greek marriage in origin
and development. There are no traces of any phase of patriarchal poly-
gamy or concubinage. The Italians were more exclusively agricultural
than the archaic Greeks, and primitive Italian marriage relations point
to loose matrilocal associations approaching clan promiscuity, in which
the husband and father was of little account, such as appear to have
survived until a later period among the Etruscans. No doubt where the

1 Some of the hetairai, at least, were remarkably cultured. Leontion, the com-
panion of Epicurus, could hold her own against Theophrastos in written
philosophical disputation, and the purity of her Attic style is praised; Thais,
Diotima, Thargelia were celebrated as philosophical disputants. The intellectual
eminence of Aspasia is familiar. The famous ‘funeral oration’ of Perikles was said
to have been composed by her; she opened a school for young women at which
she herself taught. The relations of hetairai to their lovers could be sincere and
unmercenary. The hetairai in the comedies of Plautus, though in Roman guise,
are doubtless copied from Greek models; they spurn proferred wealth in their
self-respect and fidelity to their lovers.

THE SOCIAL EVOLUTION OF MONOGAMIC MARRIAGE 23I

estates or titles of an heiress were at stake, the husband acquired them
through a more formal matrilocal association, much as with archaic
Greek princes. But, while Greek marriage appears to have evolved
directly out of such a matrilocal association and remained to the
last juridically matrilocal, in Roman usages everything points to a
sudden revolution. The wife was transferred by a legal act from the
jurisdiction of her father to that of her husband. This innovation was
introduced in a society still essentially matriarchal; matriarchal influ-
ence, therefore, instead of gradually disappearing, as in Greece, was
preserved by the very process which ostensibly abolished it, much as
pagan temples have been preserved by being converted into Christian
churches. Thus it is that in the emphatically patriarchal society of
Rome women retained a much higher position than in the far less
patriarchal society of Greece; and that the cleavage between the two
contrasted functions of the woman, as wife and as ‘hetaira’, never
became complete; thus arose the European conception in which mar-
riage is identified with sexual mating, ‘legality’ with sentiment — in
short, the fully-developed monogamic ideal.

Patriarchal Roman marriage was instituted by the patricians for their
own purposes, much as the aristocratic class among the Yurok Indians
remodelled the marriage institutions of their tribe to protect their
interests. The patrician privilege consisted in having a legally recognized
heir. The propertied patricians did not recognize the marriage arrange-
ments of the propertyless plebians as being marriage at all, since the
plebians ‘did not know their own fathers’. Moreover, they refused to
allow them to adopt legal marriage because they would thereby have
become patricians, recognized owners of property with the right to
transmit it, not to the clan, to which it went under the old law, but to
their heir, to their ‘family’.

Patrician marriage was deliberately evolved. Three forms of patri-
archal marriage were current in Rome in earlier historical times: the
most primitive was ‘usus’, by which, if a man and a woman had lived
together connubially for a year, the marriage was recognized as ‘legal’ —
that is, the children could inherit; this was orginally the only form of
marriage. Second, ‘coemptio’, a ‘legal’ union, the contract of which had
to be witnessed by five Roman citizens. The third form was in early
times the specifically patrician marriage; the contract was witnessed by
ten Roman citizens, and sealed by sacrificing a sheep. It was known as
‘farreo’, or ‘confarreatio’, because the ceremony included the sharing of
a cake of ‘far’, the common Latin flour — our wedding-cake, in fact —
consecrated by a flamen. The flamen, however, did not take any actual
part in the ceremony.

It has been suggested that the difference between ‘farreo’ and other

THE MOTHERS

232

forms of Roman marriage lay in the former being a ‘religious marriage’.
We shall see later what a ‘religious marriage’, in the proper sense of the
expression, a ‘hicros gamos’, really means; it means the marriage of a
woman with a god, and in many stages of culture it was thought incum-
bent upon every woman to go through such a marriage.

But the Romans no more thought of marriage as a sacrament in the
Christian sense than a modern bride thinks she is marrying a god. It is
true that an attenuated relic of ‘religious marriage’ was celebrated at the
same time as the Roman ceremonial, whichever of the three forms it
took. The bride performed a most indelicate ceremony in connection
with the statue of a god, the purpose of which was solely to make the
marriage contract as binding as possible; it was a legal use of religion
identical with the administration of an oath.

What was the contract which it was sought to make so binding ? It was
not the sanctification of the union, it was not an oath of fidelity. It was
simply this — that the woman passed from the guardianship, or ‘potes-
tas’, of her father to the guardianship of her husband. It was a deed of
transfer. The Roman family rested on the notion of ‘patria potestas’, the
absolute power of the father over the family. ‘Family’ comes from the
Oscan ‘famel’, a servant, slave, a possession; and ‘father’, ‘pater’, means
owner, possessor, master. The Roman ‘pater familias’ was thus literally
an ‘owner of slaves’. He had (theoretically) absolute power over his posses-
sions, his chattels, his children. He could kill them or take them to the
market and sell them. Accordingly, in Roman law the wife was tech-
nically her husband’s ‘daughter’. The contract was a consequence of the
theory of ‘patria potestas’. The transference of the woman from one
‘familia’ to another was called marriage with ‘manus’. In later Roman
legislation all three forms of marriage implied ‘manus’, but in their
original forms neither ‘usus’ nor ‘coemptio’ did so.

‘Coemptio’ was a more archaic form of marriage; as the word indi-
cates, it is a derivative of marriage by purchase. It was interpreted by
Roman jurists as a ‘mutual purchase’. Probably, as in all savage societies,
the patrilocal family, the ‘familia’, first emerged when sufficient induce-
ment was offered to the wife’s parents to allow her to leave the home,
i.e. by purchasing her. Subsequently, the patriarchal right was made
more exclusive by the ‘farreo’, said to have been introduced by Numa,
the Sabine king.

As soon as the legal distinctions between patrician and plebeian rights
of property ceased to exist, all these forms of marriage fell into disuse.
By the time of Cicero, and even as early as the second Punic war,
‘farreo’ and ‘coemptio’ were wholly obsolete. ‘Farreo’ became a mere
opportunity for occasional ceremonial display. Throughout the later
Republic and the Empire marriages were contracted, as of old, by

THE SOCIAL EVOLUTION OF MONOGAMIC MARRIAGE 233

‘usus’ only. In Roman law the proof of marriage consisted in the deeds
of dotal transfer; there was no ceremony.

In legal theory the woman who passed from the ‘patria potestas’ of
her father into that of her husband was in a position of absolute sub-
jection. ‘If’, said Cato, ‘thou findest thy wife in adultery, thou art free
to kill her without trial, and canst not be punished. If, on the other
hand, thou committest adultery, she durst not, and she has no right to,
so much as lay a finger on thee.’ In the early days a law had to be passed
making it illegal for a man to sell his wife. A woman was, in law, a
perpetual minor. She could not own or transmit property or enter into
any business transactions; even her children were not legally hers, she
had no rights over them.

In practice, however, the position of women in Rome was from first to
last more independent and dignified than among any other patriarchal
people. Precisely because the law was so artificial and an innovation, the
practice was very different from the theory. The matrimonial social
constitution of Rome presented, in fact, a striking paradox. Compared
with the Greek wife, the Roman woman was a free woman and a queen.
Roman girls received the same education as boys, and in ‘mixed schools’.
There was no Oriental seclusion, although it was not thought becoming
that she should resort too frequently to places of amusement. When her
husband entertained, she acted as hostess. The Romans themselves
noted the contrast. ‘What Roman’, exclaims Cornelius Nepos, ‘would
be ashamed to bring his wife in to dinner, and who amongst us does not
regard the mother as occupying the first place in the house and in our
regard? They do things very differently in Greece, for a woman is
never present at a dinner, unless it be among her own relations, and she
never sits down, except in the internal apartments.’

The explanation is that the matriarchal organization could not be
swept away by a legal measure. The principle of ‘manus’ became abso-
lute in the later times, simply because women would have none of it;
and the law had to give way much as in our own day women object to
promise obedience to their husbands.

Roman marriage was the last step but one in the evolution of our own
institutions and conceptions. Roman marriage was converted by Chris-
tianity from a civil contract into a religious, a sentimental act, a sacra-
ment. In Christian marriage the extraneous aspects of the relation
between the sexes, economic and sexual, have for the first time become
combined in one institution. Although Roman monogamic marriage is
socially and juridically identical with the institution of marriage as it
exists amongst us today, yet the sentiments with which it is viewed are
the products of a momentous change, effected solely by the Christian
religion. Later we shall sketch the development of these sentiments.
t.m. — 8 *

CHAPTER 16

Tabu

Primitive Ethics

Man is a moral animal — the only moral animal. Among other animals
there is no voluntary suppression of impulse and no self-reproach in
yielding to it. In order that such a conflict of motives should exist, those
motives must be consciously perceived. The natural instinct must be
checked by the force of a conscious veto.

Curiously enough, these distinctively human prohibitions, instead of
exhibiting a gradual development with the evolution of human society,
are found in the greatest abundance in the most primitive phases of
society. Every act of primitive man is burdened with prohibitions which
would be accounted intolerable by civilized man. But remarkably enough,
few of these countless prohibitions refer to what we should regard as
moral values.

For example, we consider that nothing is more to be condemned than
murder, especially the murder of a relative. Most primitive peoples,
however, view murder with strange indifference. Speaking of savages
generally, Stcinmetz remarks that ‘the only reproach which the slayer
of a blood-relation incurs is that he has hurt himself by weakening his
own family’. The injury inflicted by a murderer is held to be fully com-
pensated if the loss which he has caused to a given family is made good
by presenting the relatives of the murdered person with a substitute.
Thus among the North American Indians the feelings of a mother,
whose son had been brutally murdered, were assuaged by her adopting
the murderer in his place. Similarly, a widow whose husband had been
murdered might be consoled by marrying the murderer. Again, among
the Habe hill-men of the French Sudan a murderer supplies the family
of the victim with a woman from his own family ; when she bears a son,
the boy is given the name of the murdered man and all is well. (The
principle that the shedding of blood constitutes a loss which must be
compensated is carried to its logical conclusion by the Goajiros, among
whom a man who hurts himself is required to compensate his relatives
on the grounds that he has shed the blood of the family.)

Frequently little or no penalty attaches to murder. Among most
primitive people the murderer of a fellow clansman is regarded as
unclean, and is avoided or even expelled from the community; however,

TABU

235

as Sir James Frazer has shown, this has nothing to do with moral
condemnation, but arises from dread of the murdered man’s angry
ghost. In the island of Futuna ‘in heathen days men-slayers were usually
respected as well as feared’. Among the Fuegians a parricide is simply
sent to Coventry. Among the Nandi, a man who has murdered a mem-
ber of the clan is regarded as unclean — until he has murdered two
members of another clan, when his moral purity is completely restored.
In the Jewish code there is no definite penalty attached to homicide;
homicide may be outlawed, but punishment, or rather vengeance, is
entirely the concern of the victim’s nearest relative.

There are two main reasons for this neglect of social principles. In the
first place, social regulations are scarcely needed in primitive society.
Human society arose as the outcome of the natural instinct of solidarity
between members of the same brood. The individuality of each was
merged in the solidary consciousness of the collective group-instinct.
The individual did not defend himself as an individual; the group
defended itself as a group. Being spontaneous, this sentiment did not
need to be formulated as a law. All primitive societies are highly moral
compared with ours in respect of just those rules which in our societies are
enforced by laws and principles. Of the Marquesans, Mr Melville
writes : ‘During the time I lived among the Typees, no one was ever put
upon his trial for any offence against the public. Everything went on in
the valley with a harmony and smoothness unparalleled, I will venture
to assert, in the most select, refined and pious associations of mortals in
Christendom. There was not a padlock in the valley nor anything that
answered the purpose of one. ... I will frankly declare that after passing
a few weeks in this valley of the Marquesas, I formed a higher estimate
of human nature than I had before entertained.’ Father Veniaminoff
gives a similar account of the social morality of the Aleuts. He fancied
himself among a community of primitive Christians. Their sexual
morality, like that of the Marquesans, was of course scandalous. In fifty
years the courts of law established by the Russian government could
find no employment. In Hawaii, during the time that Lisiansky spent
there, only one man was tried and executed, and that for eating a coco-
nut on a tabu day. Of the Iroquois, a Dutch colonist, who gives a
shocking account of their sexual morals, says of their civic morality,
‘although they are so cruel, and have no laws and punishments, yet there
are not half so many villainies and murders committed amongst them
as among Christians’. Such illustrations might be multiplied. Only when
the primitive constitution of society becomes sapped by the establish-
ment of private property, and the instincts on which it is based are
consequently supplanted by individualism, do social laws and formula-
tions of moral principles become necessary.

THE MOTHERS

236

Another reason why such laws are absent in primitive society is that
such offences are regarded as a private concern. It is for the offended
party or his relatives to take revenge, and the man who slays his father’s
murderer is congratulated by all. Primitive legislation deals only with
group interests. It is in much more advanced stages that the right of
personal revenge is delegated to the state, the king or the god.

Nature of Tabu and Sacredness

The moral character of ethical principles and their character as cate-
gorical imperatives have been imparted to them by their assimilation to
primitive tabus.

When a savage is questioned as to the origin of a tabu, he is usually at a
loss to explain it, and says that the thing has always been considered
tabu. Tabus are, in fact, identical with what are among us spoken of as
superstitions, such as the notions associated with spilling salt, stepping
under a ladder and so forth; superstitions arc, in fact, survivals of tabus.
They still flourish in all secluded mountain, rural and fishing popula-
tions. The inhabitants of the Scottish Highlands, the hills of Switzer-
land, the Balkans, etc., are almost as rich in superstitions as are primitive
tribes in tabus.

Tabus are observed from dread of the consequences of neglecting
them. In more advanced stages of development the notion of tabu has
given rise to two seemingly opposed sentiments : a thing may be tabu
because it is too holy to be touched, or because it is unclean — a breach of
the tabu would pollute the offender. Thus the tabu may be an expres-
sion of extreme reverence or extreme horror. But primitive tabu con-
tains nothing of this distinction. Our word ‘sacred’, the Latin ‘sacer’, as
also the Greek word r ayQiog\ originally corresponded exactly to the
Polynesian word ‘tabu’. Sacer meant equally holy and unclean, venerable
and accursed. The bad meaning of polluting is by far the more promi-
nent in the early use of the word. ‘Sacer esto’, let him be sacred, was the
formula by which a man was outlawed, ex-communicated, e.g. for
treason or homicide. Anyone had the right to slay a ‘sacred’ man — but
I suspect that this was not the most primitive form of the rule, and that,
originally, as among certain primitive tribes, no one would have dared
kill him. Yahweh dealt with the first murderer, Cain, in this way, saying:
‘Whosoever slayeth Cain, vengeance shall be taken on him sevenfold.’

In popular sentiment this idea persists. A lock of hair, a bone, or any
part of a condemned criminal is a talisman possessing miraculous vir-
tues. In Cornwall it was believed quite lately that a criminal who was
going to be hanged could cure diseases by touching people. In Fran-
conia a salve prepared from the fat of an executed criminal possessed
such valuable properties that chemists had to put up a preparation to

TABU

237

meet the demand. At public executions it was difficult to restrain the
spectators from dipping their handkerchiefs in the blood of the victims ;
late in the last century in Berlin, men drank the blood of an executed
malefactor, believing it a remedy for all ills. The reason undoubtedly is
that a condemned criminal is a ‘sacred’ person. The execution of a
criminal was, in ancient times, a solemn sacrifice to the gods. The Celts
habitually sacrificed criminals, ‘and if a man was cannibal enough to eat
a bit of the victim’s flesh, he by that act rose in the goodwill of heaven.
He was supposed to have absorbed into his system much of the sub-
stance of what was consecrated to God.’

So completely did the good meaning of ‘sacer’ displace the bad that
the Romans were puzzled by the hieratic use of the word. But primitive
gods are not good; they are dreadful and dreaded, and worship of them
consists chiefly in placating them. Thus they are sacred in the same
sense as all tabu beings and things are sacred : the less one has to do with
them the better. When Protestant missionaries tried to explain to the
Dakotan Indians that the Bible and the Church were holy, they used the
Dakotan word for tabu, ‘wakan’. The Dakotas’ natural reply was that, in
the circumstances, they would rather have nothing to do with them. The
ancient Jews themselves took much the same view of their holy books. A
man had to purify himself and wash his hands after handling the Bible.

In short, no ethical value originally attaches to what is tabu. The
differentiation which later takes place is not an unfolding of latent ideas
but an addition. The notion of tabu, though it has come to be connected
with gods, is not necessarily dependent upon the notion of God. Primi-
tive gods are tabu and render anything pertaining to them tabu because
they are dreadful, but a thing may be dreadful without being divine.
The notion that a thing is tabu because it is the property of a god, or of
the ghost of a deceased person, implies a clear idea of personal property,
and it is therefore highly improbable that such a notion was the origin
of tabu prohibitions. Tabus could not originally have been imposed by
the authority of rulers and gods, for they are much older than such
conceptions ; in no primitive society is there a ruler invested with any
degree of power, and in many primitive societies the authority of gods is
hazy.

Different tabus have of course had various origins, but what calls for
explanation is the whole notion of the origin of a prohibition from human
instincts. Such prohibitions must have been imposed in the first instance
in a very categoric form. The frequently-made assumption that they
were imposed by the arbitary authority of chiefs is an untenable view,
for there is nothing more foreign to the character of primitive man than
respect for arbitrary authority, and truly primitive societies are nothing
if not equalitarian. It is only at the human level, through the medium of

THE MOTHERS

238

language, that a prohibition can acquire the status of a principle. If I
am right in considering that the authority of the female was paramount in
the earliest human groups, a prohibition of this kind must have been the
very first to come into operation . 1

The Tabu on Menstrual Women

The most fundamental of all primitive tabus are those referring to
women and sexual intercourse. Tabus on the approach of men to women
during and after parturition, and during menstruation, are the most
strictly observed of all primitive tabus. All the world over, and not only
among savages, the tabus attaching to menstrual women are similar.
And those which refer to women in childbed are practically identical
with them. The latter have little reference to the child, but refer to the
lochial discharge. Premature births and all issues of blood are treated in
the same manner as full-time births as regards tabu.

It would weary the reader to recapitulate the vast array of evidence
from societies in every part of the world which can be brought together
to demonstrate the similarity of such tabus. The practices of the Eskimo
may be taken as representative of all others. Four weeks before her con-
finement a woman retires to a separate hut, which no man is allowed to
approach. She remains there for a month after she has been delivered,
when the father is allowed to see his child for the first time. Should a
woman be unexpectedly seized with labour pains, all the men stop work
and build a snow-hut as quickly as possible. During her period of
seclusion the woman cannot eat or drink in the open air; even the
remains of her meals must be disposed of inside her hut. She has her
own cups and dishes, which men must be careful not to use, and is sub-
ject to dietary regulations. If the few kinds of food which are permitted
are unobtainable, she may have to go without eating. Among the Tlin-
kit, females are driven out of the house at the time of childbirth, and
only some time after birth is a mother allowed to enter a rude shelter
erected for the purpose.

In the same way, a menstruating woman is regarded as a being with
whom all contact, however innocent, would entail dreadful consequences.
In most societies a special hut is provided to which women must resort
during their menstrual period. Special ceremonies regulate the first
menstruation. Thus among the Dene a girl, at her first menstruation,

1 There is one relation in which normal impulses are interdicted — the beha-
viour included under the term ‘coyness’. However, the attribution of feminine
diplomacy to lower animals is little short of absurd. The coyness of the female
animal towards the courting male is the result simply of physiological conditions.
The female is not prepared for impregnation except at stated periods of ovula-
tion and rut. At other times the advances of the male arc not desired. Among
animals the male can only be repulsed or avoided by resistance or escape.

TABU 239

remains isolated for two lunar months. She must not touch even her
own food with her hands, and is provided with a stick to scratch her
head. In some cases the woman’s head is wrapped up so that she can see
nobody, and she is submitted to purges and fastings. In other cases a
girl, at her first menstruation, is placed in a hammock near the smoke-
vent of her hut so as to be thoroughly fumigated, and only at night may
she rise to cook a little food for herself. The vessels which she has used
are immediately broken and buried. Among the Guaranis and other
tribes the girls are sewn up in their hammocks in the same way as corpses,
only the smallest opening being left to allow them to breathe, and are
suspended over the fire for several days. Not infrequently the unfortu-
nate girls die under the process of disinfection. In some cases the women
are beaten and all their hair is plucked out. Some observers have thought
that this retiring of the women was due to an instinctive sentiment of
modesty, but since, in some cases, the women are allowed to deliver their
children upon their knees in the public street in the presence of every-
one, this is hardly likely. In Russia, a woman after delivery is regarded as
being in a state of impurity, and may not hold any communication with
others until she has been purified by a priest. In the province of Smo-
lensk she is confined in a barn or hut some distance from the house.

The Hebrews attached the greatest importance to the primitive tabu.
‘If a woman have an issue and her issue in her flesh be blood, she shall
be put apart seven days ; and whosoever toucheth her shall be unclean
until even.’ Anyone who touched her bed or anything that she sat upon
was required to ‘wash his clothes, and bathe himself in water, and be
unclean until the even’. On the eighth day after, she had to take two young
pigeons to the tabernacle, one as a sin-offering and the other as a burnt-
offering, and the priest had to make atonement for her.

In ancient Persia the very glance from the eye of a menstruous woman
was so polluting that she must not look upon a fire or upon water, nor
converse with any man. She was confined during menstruation to an
isolated portion of the house, no fire was to be kindled and everyone was
to remain at least fifteen paces from any fire or water. When the period
was over, the clothes which she had been wearing must be destroyed,
and she must be purified by being washed with bull’s urine. The tabus
attaching to child-birth were exactly similar. The rule mentioned by
Herodotus that a man might not see his own child until it was five years
old probably refers to a similar lengthy exclusion of a man from the
house of his wife. The modern Parsees still observe most of the rules of
their ancestors, and it is not long since there were in Parsee communi-
ties public menstrual houses to which women resorted at their periods.

The Hindus take as serious a view of the impurity of a menstrual
woman as did the Persians, while the aboriginal races of India segregate

THE MOTHERS

24O

menstruating women even more rigorously than the Arya Hindus. Thus
among the Kamar, when a woman is menstruating, no man of the same
household can enter a temple without previously having had a bath.

Much the same is true among wild races of the Malay Peninsula.
In Africa, if an Akikuyu woman menstruates in a hut, it is at once
destroyed as unfit for human habitation. Among the Menangkabau of
Sumatra if a menstruating woman were to go near a rice-field the crop
would fail. Among the Visayans of the Philippine Islands when a
woman is overtaken by child-birth everything is removed from the
house, else the weapons would no longer be efficient, the nets would
catch no more fish, and fighting-cocks, their most valued possession,
would no longer be able to fight. Among the Caroline Islanders, women
are wrapped up in mats from head to foot during the whole duration
of pregnancy, and seclusion continues for five or six months after
delivery. They may not paint their faces nor anoint their hair, and are
unfit to undertake any cooking. Were a man to touch so much as a drop
of water from the pools in which they wash, it would be impossible for
him to catch any fish. In Nauru, one of the Gilbert Islands, not only
may no food be eaten that has been touched by a tabu woman, but it is
considered that the nuts on any trees within a hundred feet of such a
woman’s dwelling are unfit for human consumption.

These ideas of the defiling and dangerous character of the menstruat-
ing or child-bearing woman appear in many forms. Thus in New
Britain when a birth has taken place all the men must be disinfected.
They subject themselves to fumigation and rinse out their mouths with
ginger. Women, it would seem, can infect even themselves, for among
the Pennefather and other Australian tribes a menstruating girl is buried
up to her waist in the sand, a fence of brushwood is built round her, and
she is provided with a stick to scratch herself, as she must not touch her
own body with her hands. The Wakelbura tribe of south-eastern
Australia believes that if a male should see a menstruating woman he
would die. Sir George Grey observes that the more southerly tribes of
Western Australia conform to the injunctions in Leviticus (xv. 19).

As these instances show, the tabus are not derived from any abstract
principle, but from dread of the dire effects of contact with women in
these conditions. The Dene believe that contact with menstrual blood
will turn a man into a woman. The Tlinkit are persuaded that a single
look from a menstruating woman would completely destroy the luck of
a hunter, fisherman or gambler, and that it might even, like the Medusa’s
head, turn objects into stone; a view analogous to that of the Bushman,
who believes that her glance can cause men to become fixed in whatever
position they may happen to be, and to turn into trees. Similarly the
Orang Belenda believe that contact with a menstruous woman will

TABU

24I

deprive a man of his manhood; while some South African tribes believe
that if a man should touch a woman during the period his bones become
soft and he cannot in future take place in any manly exercise. Not only
men but their weapons and implements lose their virtues. An Indian
woman who inadvertently stepped over a man’s gun while she was
menstruating had to flee for her life, and was only accepted back when it
was discovered, to the astonishment of all, that the gun could still
shoot.

Similar views obtained in primitive Greece; and sceptical as were the
Greeks of a later age, they still regarded the impurity of a lochial woman
and of a corpse as equal. Of the ideas of the Romans, Pliny has given us
a full account. ‘Hardly can there be found’, he says, in Philemon Hol-
land’s rendering, ‘a thing more monstrous than is that fluxe and course
of theirs. For if during the time of their sicknesse, they happen to
approach or goe over a vessell of wine, bee it never so new, it will
presently soure; if they touch any standing corn in the field, it will
wither and come to no good. Also, let them in this estate handle any
grasses, they will die upon it; the hcarbes and young buds in a garden,
if they doe but passe by, will catch a blast, and burne away to nothing.
Sit they upon or under trees whiles they are in this case, the fruit which
hangeth upon them will fall. Doe they but see themselves in a looking
glasse, the cleare brightnesse thereof tumeth into dimnesse, upon their
very sight. Look they upon a sword, knife, or any edged toole, be it never
so bright, it waxeth duskish, so doth also the lively hue of yvorie. The
very bees in the hive die. Yron and steele presently take rust, yea, and
brasse likewise, with a filthic, strong, and poysoned stynke, if they lay
but hand thereupon. If dogs chance to tast women fleures, they runne
mad therewith: and if they bite anything afterwards, they leave behind
them such a venome, that the wounds are incurable.’

These beliefs are substantially current among most rural populations
of Europe. In the wine districts of Bordeaux and the Rhine, menstruat-
ing women are strictly forbidden to approach the vats lest the wine
should turn to vinegar, and I have seen the same rule observed in the
Chianti district. In northern France they are excluded from sugar-
refineries when sugar is cooking, and in Holstein no menstruating woman
attempts to make butter. No Frenchwoman would attempt to make a
mayonnaise sauce while in that state. In England bacon cannot be cured
by a menstruating woman. ‘It is a very prevalent belief among females,
both rich and poor,’ writes a medical man in 1878, ‘that in curing hams,
women should not rub the legs of pork with the brine-pickle at the time
they are menstruating.’ They even render bright mirrors dim.

The importance of a woman being properly ‘churched’ after her con-
finement is therefore manifest. Pennant records the practice in his Tour

THE MOTHERS

242

in Scotland. One would scarcely expect to find in Europe the rule, so
common among savages, that a puerperal woman must not touch her
own head. Yet in Bavaria it is thought that if a woman in child-bed were
to comb her hair it would turn grey.

Sexual Separation during Pregnancy and Suckling

Such tabus are not the only restrictions on sexual relations in primitive
societies, for it is the general rule that all cohabitation must cease when
a woman becomes pregnant, or at any rate during the later months of
pregnancy, and this commonly continues during the whole time she is
nursing a child, a period much longer among uncultured peoples than
with Europeans. Children practically wean themselves after anything
from four to seven years of suckling, though from two to three years is
the most general duration. 1

Restrictions of this kind are reported from all over the world. Thus
among the Warega a wife will not allow her husband to approach her
until the child can walk, though he may visit her if he sits quietly at the
opposite side of the hut. Such restrictions are general among most
African tribes. Among the Suk of Kenya Colony no marital relations take
place until the child has cut two teeth.

It is reasonable to believe that such regulations were at one time even
more general in primitive society than at the present day. The restric-
tions on the instincts of the male in the more primitive societies are
evidently much more extensive than any observed by civilized Euro-
peans. Even the tabus upon menstruating puerperal women are not
regarded by Europeans with the same dread as attaches to them in
uncultured societies. Although the Synod of Wurzburg reiterated in
1298 the prohibition that ‘no one shall approach a woman who is near
her confinement or in her menstrual fluxes’, even theological opinion
held such an act to be but a venial sin. Ecclesiastical exhortations during
the Middle Ages show that abstinence at such times was by no means
scrupulously observed. Among uncultured peoples today these customs
are tending to become relaxed, and it is usually enough that a menstruat-
ing woman should give proper warning of her condition by such acts as
painting her face yellow (as among the Mandingo), or wearing a brightly
coloured scarf.

Origin of Periodical Sexual Prohibitions

This tendency of such tabus to fall into disuse appears inconsistent with
the supposition that they first arose as a spontaneous inhibition of
masculine sexual instincts by feelings of repulsion, since it implies a

1 A bibliography of 31 titles is appended.

TABU

243

greater degree of squeamishness in savage than in civilized man. It has
commonly been assumed that such tabus originated in some feeling of
disgust. But primitive man is not disgusted by those things that Euro-
peans regard as repulsive. In the matter of food, there is nothing too
disgusting for savages to eat. Some use their cesspools as reserve stores.
Primitive men exhibit no greater delicacy in regard to the satisfaction of
their sexual instincts. Those savages who are most scrupulous in observ-
ing the menstrual tabus will also perform acts the very recital of which
to us is nauseating. Australian aborigines who die of fright because a
menstruating woman has touched their blanket habitually perform
homosexual atrocities unknown to European vice.

When a savage manifests disgust, it is almost invariably at the breach
of some tabu, such as eating some food, maybe quite appetising, which
is prohibited by his customs. The tabu is the cause, not the effect, of his
disgust.

The awful character of the menstruating woman has been set down
to a supposed primitive ‘horror of blood’, notably by M. Durkheim,
but there is no evidence of any such horror — blood is everywhere re-
garded as a delicacy . 1 Even peoples whose habitual diet is entirely
vegetarian will miss no opportunity of indulging in a daught of warm
blood. The Masai have a special method of shooting a blocked arrow
into the jugular vein of an ox, so that they may drink the warm blood
before they slaughter it. Sir Harry Johnston describes how, after filling
their wooden vessels, they step apart from the crowd ‘to drink the
coagulating gore with utter satisfaction and a gourmet’s joy’. The
Australian natives, besides drinking one another’s blood on every
occasion, use it to bedaub themselves and to attach feathers and orna-
ments to their bodies. There is no instance known of blood in general
being regarded with horror by any uncultured peoples.

It is frequently supposed that the menstruating woman is possessed
of some dangerous spirit, and the fumigations and flagellations which
we have noted are intended to expel this spirit.

That so many speculations should have been put forward is surpris-
ing, for these tabus are the only prohibitions which are not peculiar to
mankind, but are common to all mammals. Among all animals the
female admits the male only at such times as she is prepared for the
exercise of his function; at others, her attitude to him is one of positive
hostility. The description given by Pallas of the way the female camel
rounds on the bull the moment she is impregnated, driving him away
with snarling, is representative of the behaviour usual with mammalian

1 There are special circumstances in which the blood of a particular person
may, like any other substance, be the object of a tabu. It is not blood in general
which is fraught with dread, but the blood of women.

THE MOTHERS

244

females. Had the primitive human female admitted the male during
menstruation, pregnancy and lactation, her behaviour would have con-
stituted a biological abnormality.

As in all physiological functions, there is an adjustment between the
inherited disposition of the male and that of the female. It has been too
readily assumed by naturalists that all animals must have a special
breeding season, but our information, fragmentary as it is, suggests that
among most, if not all, carnivores the time of year at which the young are
bom varies, in the same species, within wide limits. In contrast to
herbivores, there is no indication that at any time the sexual instincts of
the male are completely quiescent. Where there is no definite yearly
periodicity in the female, there can be no exact adjustment of the male
instincts to that function. ‘Among numerous higher species the males
appear to be at all times disposed for sexual union, even though the
females are not in rut. 5 In the primates the mutually-adjusted periodicity
of sexual instincts has entirely disappeared. To regard these facts as
imposing restrictions on sexual intercourse in primitive humanity is not
a matter of hypothesis, it is inevitable. Those restrictions are imposed
by biological necessities, and are in every respect the same as those
obtaining among animals. As with every such traditional prescription,
all manner of superstitious interpretations have become attached to
the categorical prohibition. They are manifestly accretions on the ori-
ginal veto laid by women on the sexual instincts of the male. The
formulated prohibition, like the biological restriction, must have been
imposed by women upon men and not the reverse; nor are they restric-
tions imposed on themselves by the men.

It is often suggested that the restrictions upon sexual intercourse
during pregnancy and lactation are dictated by concern for the welfare
of the child. Sometimes the seclusion to which the woman is subjected is
similarly explained, but these are manifestly attempts to account for the
custom, and are of little importance as indicating its origin.

The terms in which many accounts are couched suggest that these
observances are imposed upon the women by the tyranny and super-
stition of the men: the women are ‘compelled’ or ‘driven out’. But there
is little evidence that any compulsion is needed to force the women to
segregate themselves at such times, and probably the wording merely
expresses the assumption of the reporters that any ethical regulations
must have been imposed by the men, except where such observances
are set down to feelings of ‘natural modesty’ on the part of the women.
Even where the men are most tyrannical, the women never carry out
their arduous duties under compulsion, and similar tabus in their most
rigorous form are observed in societies in which the women, far from
being tyrannized, exercise an almost despotic power over the men. The

TABU 245

women, in most accounts, segregate themselves of their own accord,
without consulting the men.

It has sometimes been remarked that such customs are abused by the
women as a pretext to get rid of their husbands. Speaking of the Beaver
Indians, Mr Keith says, that a woman ‘pretends to be ten days in this
state and suffers not her husband except upon particularly good terms.
Her paramours, however, are permitted to approach her sooner\

Papuan tradition represents the men as quite enraged at their being
repulsed by their menstruating wives. The women of the Tully River
district in Queensland assert that they are anxious to menstruate regu-
larly, for if they did not ‘the men would be enabled to continually pay
them sexual attentions, a course to which the women assured me they
objected’. In the mythical Alcheringa days, when the tribesmen were
celebrating the ceremonies of the bandicoot, their totem, a woman,
through excessive repetition of the rites of fertility, began to bleed pro-
fusely. She said: ‘I will no longer be a woman, but a bandicoot; and you
cannot touch me.’ And ever since then the women have had monthly
periods, during which they are tabu.

The awful but undefined dangers which attend the breach of a tabu
are identical with the dangers which result from a curse, than which
nothing is more dreadful; and no curse is more dreaded than a woman’s.
Among the Damaras, if a woman curses her husband, he cannot, what-
ever may happen, cohabit with her again.

That curse has recoiled upon woman herself. The notions of her
impurity which pervades the ideas not only of the savage but also of
more advanced peoples, have their root in the primitive menstrual tabu.
Woman is ‘the cause of all evil’. All tabus fraught with vague dangers
tend to become extended in their application. Thus the prohibited
degrees of marriage have sometimes come to include the very persons
who were originally prescribed as the natural sexual mates, such as
cousins and wife’s sisters. The tabu of modesty which originally applied
only to the sexual organs has often come to include the whole body. In
the same way the precautions taken with regard to the menstrual or
parturient woman have frequently been extended to women at all times.
In China the same care is used in handing anything to a woman as in
Alaska in passing provisions to her when she is in quarantine. In Tahiti
women were excluded from religious festivals, neither could they at
any time eat in the company of a man, nor even sleep in the same
house.

But it does not follow that because women were hedged about with
such vexatious tabus that their position was one of degradation.
Throughout Polynesia the position of women was the reverse of de-
graded; yet this is the country where the observance of tabus reached

THE MOTHERS

246

its most extravagant development. In Hawaii, since a woman could not
eat in the company of her husband, and neither could he eat anything
which she had cooked, there was no alternative for him but to do his
own cooking, and he cooked for her as well. Thus the tabu results in the
women being waited on by the men. Similarly, since it would be danger-
ous for the woman to handle fruits and vegetables intended for food, the
men felt impelled to take over the agricultural work, while the women
had nothing better to do than to amuse themselves making gorgeous
dresses. When, therefore, Mr Ellis and other travellers deduce the
shamefully-degraded position of women from the fact that they may not
eat with the men, they mistake the nature of tabu. An instance of the
misapprehension to which observers who do not understand the nature
of tabu is liable is afforded by a recent traveller in North Africa, who
considered that no further proof of the degraded position of women
among the polygamous Arabs is needed than that Arabs place women
and pigs in the same class. But, of course, both women and pigs are
tabu, which in primitive conceptions may mean either sacred or im-
pure. In this sense, the ancient Jews placed both pigs and Yahweh ‘in
the same class’.

The tabus attaching to women were not originally identified with such
ideas as impurity, though they have commonly become associated with
them. Originally, they sometimes had the opposite meaning. In many
primitive rites the sexual act is a sanctifying and purifying ritual. In this,
as in all other instances, tabu means primarily ‘dangerous’. According
to Vedic conceptions, ‘the blood of the woman is a form of Agni, and
therefore no one should despise it’, and the ‘Institutes of Vishnu’ lay
down that to kill a menstruating woman is as great a crime as to murder
a Brahman. In some instances menstrual and lochial blood has a sancti-
fying and purifying influence. Among the Ainu menstrual blood is a
talisman. When a man sees some on the floor of a hut, he rubs it over
his breast believing he will thereby secure success. Although the Dene
regard menstruating women with extravagant dread, if a child is not
thriving his mother will fasten round his neck a piece of cloth soiled
with menstrual blood. In such instances, the menstrual or lochial blood
is, in all probability, regarded as scaring away the evil spirits and
influences, rather than as communicating a positive blessing. By this
principle the dreaded properties of the blood are often turned to a use-
ful purpose. Among the North American Indians, when the com began
to ripen, a woman would leave her isolation-hut in the middle of the night
and walk naked through the fields, thus destroying the caterpillars. The
ancient Greeks had hit upon the same ingenious procedure, and Colu-
mella recommends the same method, which was evidently used in Italy,
where it is practised at the present day in the district of Belluno. In the

TABU

247

sixteenth century the same device was used near Nuremburg to get
rid of garden pests, and in Holland today it is usual for a girl at her
menstrual period to go round the cabbage patch in order to dispose of
the caterpillars. Northern Rhodesian natives, similarly, believe that this
will drive away the tsetse fly.

It will be seen from these instances how easily the dreaded character
of a sacred or tabu person may logically pass into the seemingly opposite
notion of a wholly beneficent influence. In primitive conceptions tilings
regarded as sacred are as often as not looked upon with horror and
repulsion, and those which have come to be conceived as impure are as
commonly regarded as divine.

The Menstrual Tabu as the Type of Prohibitions

The menstrual and puerperal tabus appear to be regarded by some as
the very type of tabu. The Polynesian word ‘tabu’, or ‘tapu’, appears to
be closely allied to ‘tupua’, signifying menstruation. The word ‘atua’,
which is usually translated ‘God’ and applies to all supernatural pheno-
mena, refers to menstruation in particular. The Dakotan word corre-
sponding to tabu is ‘wakan’, which is defined as ‘spiritual, consecrated ;
wonderful, incomprehensible; said also of women at the menstrual
period’.

The assumption that the former meaning is derived from the latter is
as probable as the reverse. Among the Arabs the expressions ‘pure’ and
‘impure’ originally referred exclusively to the condition of menstruating
women. Among the Jews the medium of purification was known as ‘the
water of separation’, a term used in reference to the menstrual seclusion
of women. The tabu state is generally signified by marking a person or
object with blood; red paint serves equally well, for the blood is not the
cause of the tabu but the mark of it. The condition of a manslayer
whose deed has rendered him ‘sacred’ is indicated by painting him
red.

The practice of painting corpses and bones red, which is well-nigh
universal in primitive society and has been so from earliest ages in
Europe, may be regarded as connected with blood as one of the forms
of the vital principle, or soul, but even this aspect is not unconnected
with menstruation, for it is from the menstrual blood retained in the
womb that human beings are believed by primitives to be formed. The
Maori expressly state that blood is the substance of the human spirit.
The condition of women in the tabu state is commonly indicated by
their painting themselves red. Thus, several Australian tribes mark
menstruating women with red paint, as do Kaffir women; while in
India such a woman wears round her neck a handkerchief stained with
menstrual blood.

THE MOTHERS

248

The sign of blood is commonly used to make women tabu in marriage
ceremonies. The theory of this is expressed by Brahmanic writers : ‘The
wife is pure to her husband and impure to every stranger.’ Marking the
bride is an essential of the marriage ceremony throughout modern
India. The parting of her hair is commonly stained with vermilion.
Among the Santals, if a young man succeeds in laying a dab of red paint
on a girl’s forehead, she is his wife. It is considered a serious offence to
do so, but the marriage cannot be dissolved. The same practice obtains
in China. The bride is smeared with blood among such peoples as the
Chuckchi, the Koryak, the natives of Borneo, of the Congo, of the
Solomon Islands and in Australia.

Men, on certain occasions, especially when embarking on some peri-
lous undertaking, paint themselves red, a practice common in Africa.
Pliny tells us that the Ethiopian nobles of his day did so, and Herodotus
noted the custom among the Libyans and Arabs. It was common among
the primitive Egyptians: cups of red ochre have been found in First
Dynasty tombs. Among the early Romans, war-chiefs were painted red,
and Camillus proceeded to the Capitol bedaubed with vermilion. The
idea, no doubt, was to scare away evil and envious spirits; just as
mothers among the Tlinkit Indians safeguard their children by painting
their noses red. The door-posts of houses are marked with blood or red
paint, especially in times of danger and epidemics. This is done in West
Africa, among the Dayaks in Bengal, as it was in ancient Peru and among
the Jews.

All sacred or tabu objects are commonly marked with blood or red
paint. The sacred stones of various native tribes in India, in Burma, in
Madagascar, are marked with blood; among the Estonians sacred trees.
The Greeks painted the statues of Dionysos red, and the Romans
applied red paint to the face of Jupiter before their festivals. The
Banyoro, on the day of the new moon, waylay a man and cut his throat
in order to smear with his blood the royal fetishes. The Australian
Blacks, who pour blood over their sacred stones, and paint themselves
red after their rites, volunteer the information that this red paint is
really the menstrual blood of women. In a Hottentot song addressed to
the spirit of rain she is addressed: ‘Thou who hast painted thy body red
like Goro; Thou who dost not drop the menses.’ The deity referred to,
there can be little doubt, is a form of the moon, and the red ochre with
which they paint themselves is called ‘gorod’ after her.

Some other Tabus

All these facts point to the conclusion that the veto imposed by women
upon masculine impulses during their periodical unfitness for sexual
functions was the earliest formulated prohibition imposed upon instinct.

TABU

249

and therefore the prototype of all such prohibitions. Actually all subse-
quent tabus tended to become assimilated to the pattern of the prototype.
In some societies the arbitrary imposition of tabus became a form of
tyranny in the hands of priests and chieftains, who imposed tabus to suit
their purpose. The most highly-developed codes of moral prohibitions
preserve the original type of such interdicts, for none are regarded as
more important than those which refer to the restraint of the sexual
instincts, and morality in the highest cultures is understood to be
synonymous with sexual morality.

In primitive society sexual restrictions are connected with nearly
every activity of primitive man — the hunter, the fisherman, the warrior,
the agriculturist, the magician, regard abstention from intercourse and
contact with women as essential to the success of their undertakings.
Other tabus, less obviously connected with the primitive prototype, are
nevertheless found on inquiry to be related to it. As illustration, we will
consider two tabus still observed among us.

Few people would consider, I suppose, that any connection exists
between the superstition that it is unlucky to step under a ladder and
the tabu on menstruating women. The unsuperstitious regard it as a
precaution against the possible presence of a paint-pot on the ladder,
but the superstition is much older than the use of paint-pots or ladders.
It is a survival of what appears to be a very widespread scruple against
passing under anything at all, a superstition directly connected with
the menstrual tabu. As Sir James Frazer says: ‘The Australian
Blacks have a dread of passing under a leaning tree, or even under the
rails of a fence. The reason they give is that a woman may have been
upon the tree or fence, and some blood from her may have fallen on it
and may fall on them’ ; and he cites instances from the Solomon Islands,
the Karens of Burma, the Siamese and others. A Maori will never lean
his back against the wall of a native house, for Maori women have a
habit of thrusting their soiled diapers through the houseboards. It is a
common notion that a woman should not step over one’s legs when one
is sitting down, for if she does one will lose one’s power of running. The
Oraons of Chota-Nagpur are horrified if they see a woman climbing on
to the roof of a hut.

The facts mentioned by Sir James Frazer are not adduced by him in
elucidation of our superstition of walking under a ladder, but of the rule
observed in Rome that the Flamen Dialis may not walk under a trellised
vine, for the juice of the grape which was commonly regarded as the
equivalent of blood might have fallen upon him. But there can, I think,
be little doubt that our own superstition is connected with the same
order of observances.

250

THE MOTHERS

The Sabbath

A much more important example of a tabu still observed among our-
selves, and one which is derived from the Hebrews, is the keeping holy,
or tabu, of the Sabbath Day.

We have seen that it is an almost universal rule that a menstruating
woman must not do any kind of work, a tabu which frequently extends
to her husband. Among the Habe, a man whose wife is menstruating
would not dare to go on a journey or hunt or sow; indeed, it would be
pointless to do so, for the undertaking could not prosper. Among the
Monumbo people of New Guinea the husband is subject to so many tabus
during his wife’s pregnancy that he is virtually a pariah. He must attend
to all his wants himself and cannot borrow so much as a light for his
pipe. An Eskimo husband is incapacitated for work for some weeks after
his wife’s confinement, and he should not enter into any commercial
transaction. Sometimes the tabu on the woman affects the entire house-
hold. Thus among the Naga tribes of Manipur even the fact that a
bitch has laid a fitter of puppies or that the cat has had kittens is sufficient
to place the whole house under an interdict. These restrictions some-
times extend to the whole community, all of whom remain at home the
next day and do no work in the fields, for no good would come of it,
owing to the influence of the lochial discharge.

General abstinence from all kinds of work is generally observed by
almost every primitive community on inauspicious occasions, such as an
eclipse, an epidemic or the death of an important personage. As Profes-
sor Webster remarks, those general tabus ‘are to be assimilated to those
which rested upon individuals alone’.

Among many peoples there also exist regular tabu days recurring at
fixed intervals. To the superstitious mind all days are distinguished into
‘lucky’ and ‘unlucky’. An old resident of Ashanti calculated that there
were only about one hundred and fifty or so days in the year in which
any important business could safely be done. The ancient Greeks were
almost as bad: Hesiod’s poem on ‘Works’ lays down on what days it is
safe to undertake farm work. The chief purpose of the ancient Egyptian
and Babylonian calendars appears to have been to mark the lucky and
unlucky days. The day of the new moon is generally looked on as
especially inauspicious, and to this the day of the full moon is generally
added. ‘According to the rules of Astrology’, observes Aubrey, ‘it is not
good to undertake any business of importance in the new moon, and not
better just at the full moon.’ To be on the safe side two more days should
be added, the four days corresponding to the four phases of the moon.
With many African peoples the day of the new moon is a day of general
abstention from work. Thus among the Baziba it is ‘a recognized day of

TABU 25I

rest’. Among the Baganda firewood must not be cut or food cooked on
that day.

The two great monthly festivals of ancient Hindu religion, on which
the sacrifice of the soma, or moon-plant, is celebrated, took place on the
days of the new moon and full moon. On these days a Brahman might
not trim his hair or beard or nails; he must not set out on a journey, nor
sell any goods. The ‘Vishnu Purana’ lays down that anyone tending to
secular affairs on these days will go to hell. These observances later
became extended to the first days of the intermediate phases of the moon,
making the tabu a weekly one. In contrast with observance of the
sabbath in English and Scottish households, where it was a strict rule
that secular literature should be put aside, the Brahmanical religious
considered that the four monthly tabu days were the very days on which
the sacred books should on no account be read, these days being in-
auspicious. The Scriptures should not be exposed to the risk of pollu-
tion. The four monthly sabbaths of Vedic and Brahmanical religion
were adopted by Buddhism, and corresponding beliefs are found in
Burma and Siam.

As we might expect, the observance of tabu days attained its most
perfect form in Polynesia. The Hawaiian sabbath had quite an old-
fashioned English aspect. ‘Men were required to abstain from their
common pursuits, and to attend to prayer morning and evening. A
general gloom and silence pervaded the whole district or island. Not a
fire or light was to be seen or canoe launched; none bathed; the mouths
of dogs were tied up, and fowls were put under calabashes or their
heads enveloped in cloth ; for no noise of man or animal must be heard.
No persons, except those who officiated at the temples, were allowed to
leave the shelter of their roofs.’ In Samoa the sabbath was similarly
observed, but there was only one regular sabbath in the month, at the
new moon. The Polynesians therefore took to the fourth commandment
like fish to water when they were instructed in the truths of the Christian
religion, but the Hawaiians complained that the missionaries were far
too lax in enforcing the sabbath, and Sunday Leagues were formed
amongst the most influential natives, who sent deputations to the clergy-
men imploring them to be stricter. The missionaries, though shocked
by the lasciviousness of the natives and by their cannibalism, felt that
people so thoroughly imbued with the principles of Sabbath observance
could not be wholly bad.

It has been argued that the fact that periodic observance is timed to
take place at a given phase of the moon does not prove that it is associated
with the moon, except in so far as the moon is a time-piece. But this
view is untenable; no distinction is drawn in the primitive mind be-
tween the relation of adventitious synchronism and that of cause and

THE MOTHERS

252

effect. You will not easily convince a savage that when two events
habitually take place at the same time* the one is not the cause of the
other. The moon, as will later be shown, is not regarded by primitives as
merely the measure of time but as the cause of time. 1 It is abundantly
clear from these examples that days of general tabu observance,
whether arising from special eventualities or whether periodic, do not
originate from belief in the benefit of rest from labour, but from the
persuasion that no business undertaken on those days could possibly
prosper. It is the moon which is regarded as exercising the dangerous,
malign influence, which is at its peak at certain phases of the lunar cycle.

Though the conceptions attaching to the influence of the moon
among primitives are complex, there can be little doubt that the original
ground for the maleficent character universally ascribed to it is its
association with the sexual functions of women. (As Darwin has sug-
gested, the actual connection between the physiological and the lunar
cycle is probably due to the fact that animal life developed originally in
the sea, and hence was subject to the periodic environmental changes
produced by the tide.) The correspondence between the cycle of sexual
functions in women and the cycle of lunar changes has been noticed by
even the rudest peoples. In every part of the world women reckon the
periodicity of those functions by the moon. Menstruation — that is,
‘moon-change’ — is commonly spoken of by all peoples as ‘the moon’.
The peasants in Germany simply refer to women’s periods as ‘the
moon’; and in France it is called ‘le moment de la lune’. I have heard of
a judge in a native court in India being puzzled by the statement that a
female witness was unable to attend the court because of the moon. The
idea that menstruation is caused by the new moon is universal. Papuans
say that a girl’s first menstruation is due to the moon having had connec-
tion with her during her sleep, and the Maoris speak of menstruation as
‘moon sickness’. A Maori stated : ‘The moon is the permanent husband,
or true husband, of all women, because women menstruate when the
moon appears. According to the knowledge of our ancestors and elders,
the marriage of man and wife is a matter of no moment ; the moon is the
real husband.’ We shall see that such conceptions are by no means
peculiar to the Maori, but pervade primitive thought.

By all peoples in lower phases of culture the moon is regarded
primarily as a male. This is doubtless due to the notion that menstruation
is due to sexual intercourse between the moon-god and women. In later
stages the sex of the lunar power is changed, and the moon becomes the
chief goddess, the Great Mother. Every great female deity, every Great
Mother — Isis, Ishtar, Demeter, Artemis, Aphrodite — has the attributes
of the moon and is a moon-goddess.

1 See pp. 295-6.

TABU

253

The cult of moon deities, whether male or female, is everywhere the
cult of women. Thus among the Ibo of Nigeria the periodical sabbath
on the day of the new moon is called ‘The Women’s Day’. In the Congo
women observe special rites at the new moon. The Wemba women
whiten their faces at this time, and the Aleutians dance in the moon-
light. These feminine lunar observances are conspicuous in the more
advanced religions of western Asia, Egypt and Europe. Relics of them
were, until lately, found in our midst. Thus in Yorkshire and the
northern counties, according to Aubrey, ‘women doe worship the new
moon on their bare knees, kneeling upon an earth-fast stone’. In Ire-
land, on first seeing the new moon, they fall on their knees, saying: ‘O
Moon! leave us as well as you found us.’

The Jewish Sabbath, as we know it, resembles that of the Babylon-
ians, which was explained as a day of abstinence or propitiation. It is
known as an ‘evil day’. In the calendar of Elul II in the British Museum,
the seventh days are marked out as days on which no work should be
undertaken. All other days are set down as favourable. Probably in
earlier times only two days a month, the new moon and the full moon,
were observed as sabbaths. We know from cuneiform tablets of the
fourth dynasty of Ur, dating from the third millennium B.C., that at that
period these two days were the chief days for sacrificial observances in
Sumer. With the early Semites, as with the ancient Hindus, the Dayaks
and the Polynesians, the monthly or fortnightly observance of tabu days
became extended to four monthly days.

It is not unlikely that on her ‘evil day’ the goddess was thought to be
actually menstruating. The notion that goddesses are subject to the
infirmities of mortal women is common in India today. Thus in Bengal
it is believed that at the first burst of rain: ‘Mother Earth prepared
herself for being fertilized by menstruating. During that time there is an
entire cessation from all ploughing, sowing and other farm work.’ The
menstruation of the Earth-goddess is thus observed by the Bengali as a
sabbath. In Travancore there is an important ceremony known as
‘trippukharattu’, or purification ceremony, in connection with the
menstruation of the goddess, which is believed to take place about eight
or ten times a year, in which a cloth wrapped around the metal image of
the goddess is found to be discoloured with red spots and is subse-
quently in demand as a holy relic. Another menstruating goddess is
found at Chunganur. It is firmly believed even in some parts of Europe
that the moon regularly menstruates. The peasants of Bavaria, for
instance, when the moon is on the wane, say that she is ‘sickening’,
using the same expression as they employ for a menstruating woman.
Since the days of Homer, it has commonly been believed that drops of
blood frequently fall from heaven, and this is commonly known among

THE MOTHERS

254

rural populations as ‘moon-blood’. In Switzerland ‘the peasants do not
look upon the “moon-blood” as a figurative expression but as an actual
physical fact, in the same way as they look upon the moon itself as a real
living being. Hence the moon is spoken of as “sickening”.’ In Ashanti
the day of the new moon is called ‘the Day of Blood’, and the Yoruba
believe that if they were to work in the fields that day the com would
turn blood-red.

The Jewish Sabbath dates from before any contact of the Hebrews
with the Babylonians. The Jews reckoned it among their oldest institu-
tions. The explanation of it as a day of rest connected with the account
of the creation was a late theory, according to most competent critics.
The division of the divine labours into six days, followed by a day of
rest, proceeds from the institution of the seventh-day sabbath, rather
than the latter from the former. The Jewish Sabbath is primarily a new
moon and full moon observance, extended later to each phase of the
moon. In a subsequent chapter we shall see how intimately ancient
Hebrew religion was associated, in its origins, with lunar cults, though
traces of this association were carefully obliterated by editors of the
sacred scriptures. Yet the connection of the Sabbath with the phases of
the moon is constantly referred to in the Old Testament. Thus when the
Shulamite woman wished to consult the prophet Elisha, her husband
asked : ‘Wherefore wilt thou go to him today ? It is not new moon or
Sabbath.’ And Amos represents the Jewish profiteers as impatient at the
restrictions placed on the business by the tabu days, and as exclaiming :
‘When will the new moon be gone, that we may sell grain? and the
Sabbath that we set forth wheat . . . ?’ Isaiah, at a time when it was still
uncertain what practices belonged to the religion of Yahweh and which
were foreign corruptions, denounced in one breath both ‘new moon and
Sabbath’. The two are thus constantly associated. (See also Hosea ii. n ;
Psalms lxxxi. 3.) This association has never been effaced. It is considered
among Jews to be ‘a very pious act to bless the moon at the close of the
Sabbath’.

As with most primitive peoples, the observance of the Hebrew
Sabbath is regarded as the particular concern of the women. Hosea,
inveighing against the corrupt practices of Jewish women and speaking
in the name of Yahweh, exclaims : ‘I will cause all her mirth to cease,
her feast days, her new moons, and her Sabbaths, and all her solemn
feasts.’ The character of the sabbath as a women’s festival is several
times referred to in Talmudic literature, and the sabbath itself is
spoken of as ‘the Queen’, or ‘the bride’. In the Kikur Sh’lh, it is set out
that ‘God has given the first day of the month as a festival more for
women than for men’. To this day, among Jews, it is customary among
women to abstain from work on the day of the new moon.

TABU

255

The Hebrew Sabbath did not differ in origin and character from the
tabu days of the most primitive peoples. The Jewish institution, already
popular in Roman society, where Oriental religious observances had
become fashionable, was adopted by the Christian Church. But Greeks
and Romans had been immemorially familiar with the observance of
restrictions on certain tabu days, and especially the day of the new moon.
When the trial of the bow of Odysseus was proposed to the suitors, they
objected that it was the day of the new moon and therefore unbeseeming.
The Greek lunar tabu days never extended to the four monthly phase
days, but were mostly confined to the first. Although there was no
general stoppage of business, most public activities except those of a
religious character were intermitted, and it was considered unsuitable
for farm-work. A less solemn monthly festival was held on the day of the
full moon. The Romans observed the Kalends, or days of the new moon,
in a similar manner, and in old-fashioned households it was customary
for the paterfamilias to remain at home and offer prayers to the family
gods. The introduction of the Hebrew usage did not, therefore, bring
any startling change to the Romans.

Thus, although our inquiry into the connection between the Sabbath
and the tabu on menstruating women may have taken us somewhat far
afield, the connection is a close one; and the observance which in our
own country bears most closely the character of a primitive tabu is
immediately dependent upon what we have reason to regard as the first
tabu, or moral prohibition, imposed upon the instincts of primitive
mankind.

CHAPTER 17

The Totem

The Sacredness of Food

Most tabus refer to food and to the manner of eating it. For instance,
the strict rules regulating food preparation among Jews and Muham-
medans are familiar. In the religion of civilized peoples the act of eating
plays a prominent part ; the chief Christian sacrament is a mystic meal,
and meals are often begun and sometimes ended with a blessing. In
Scotland neglect of these observances is considered dangerous, not only
to the consumers but to anyone near. A man who suddenly felt faint in a
Scottish village was told that he had passed a house where people had
taken food without asking for a blessing ; to counteract this, he must eat
a piece of bread over which the name of the Trinity had been pro-
nounced.

When a person is tabu, the tabu applies to everything which has been
part of his body, such as his hair or spittle ; also to food intended for him
or left over by him. The strictest menstrual prohibitions refer to men-
struating women handling or cooking food, which would thereby become
poisonous. Since these rites conflict with the utilitarian value of food,
they imply some underlying interest ; the only such interest relevant to
primitive psychology is the reproductive instinct. Just as food is the
prototype of utilitarian and economic values, the reproductive impulse is
the prototype of emotional non-utilitarian ones, and since the former are
typically male interests and the latter female and non-individualistic, it
is here that we must look for the explanation. 1

Ignorance concerning the Physiology of Generation
The knowledge that the father contributes equally with the mother to
the life of the offspring dates only from the seventeenth century, and
Spallanzani’s discovery in 1785 of the male germ-cells was greeted by
fierce controversy. In Aristotle’s view the child was formed entirely from
menstrual blood; and, as Pliny said, ‘The man’s seed serveth in steed of
runnet to gather it round into a curd which afterwards in process of time

1 The comparative sterility of women in creative art appears to contradict this
distinction, but it must be remembered that art represents an opposition between
intellectual control and emotional motive. A woman, as Madame de StaSl
remarked, either has children or writes books.

THE TOTEM 257

quickeneth and groweth to the forme of a bo die.’ This is a general view
in the lower phases of culture.

Paternity was not regarded by the most primitive peoples as a physio-
logical relation, but as a social and juridic claim. A father is ‘responsible’
for a woman’s child in an economic and juridic sense; the Roman jurists
declared that the relation between mother and child is a natural fact,
whereas paternity belongs entirely to the sphere of civil law, and this is
so not only because the actual paternity may be uncertain but because
the physiological conception of paternity is absent. The peasant popula-
tions of Europe hold similar views. Physical or mental resemblance
between father and child is accounted for in other ways, e.g. it is thought
that the child is a reincarnation of some ancestor or that the mother’s
maternal impressions impart to the child the likeness to the father.
When in the early days of Australia women gave birth to light-coloured
children in tribes near to white settlements, this was accounted for by
the changed diet, such as white bread obtained from the white man.
Spencer and Gillen state that the Australian tribes they studied recog-
nized no relation between sexual congress and reproduction, a report
which, when first published, gave rise to much incredulity, but which
has since been independently confirmed and is reported of almost all the
wilder Australian tribes.

This is not surprising, for primitive ideas of reproduction are often
extremely vague. Some tribes have been astonished to learn that men-
struation could occur without sexual intercourse; others believe that
pregnancy can only result from repeated connections. The Sinaugolo
Papuans believe that conception takes place in the breast and have no
knowledge of the female internal organs. Among savages pregnancy is
usually diagnosed by comparatively late symptoms, often not before
quickening takes place. The Baku Bahau of Central Borneo think that
pregnancy lasts from four to five months ; that is, they are ignorant of
the date of conception. The Tuareg of the Ahir think it possible that
pregnancy may last several years. Or again, in Europe, the southern
Slavs believe that it may sometimes last only six weeks. The Baku Bahau
expect to breed from castrated dogs. This is not too surprising, since
Sicilian peasants do not suspect that the seminal fluid is secreted by the
testicles, but believe it comes from the spinal marrow. In the Trobriand
Islands girls who wish to avoid pregnancy are more careful not to bathe
at high tide than to abstain from sexual intercourse. The North Ameri-
can Indians, like the Australians, believe that women may be impreg-
nated by the souls of the unborn.

In general, among uncultured peoples sexual congress is not looked
upon as the cause of conception, even where it is regarded as related to
it; a woman conceives because a spirit entered into her, and man can at
t.m. — 9

THE MOTHERS

258

most act as a medium for the transmission of that spirit. Thus the Nuer
of the Upper Nile attribute pregnancy to their god, Kosz, ‘with the
assistance of the husband’. Analogously, the Roman Catholic Church
teaches that the soul of the human being ‘is created and united by God
to the infant body yet unborn, which union is called passive conception,
in which parents have no part’. The problem how children could be born
from an adulterous union, since God is the real father of every child and
yet condemns adultery, puzzled some of the early Fathers.

Immaculate Conception through Food , etc.

Belief in immaculate conception is universal. It is the usual manner in
which gods and heroes are conceived even in non-savage societies, and
legends of impregnation by non-human agencies are innumerable. The
Subanu of Mindanao, for example, believe that conception may take
place through the operation of the moon, the sun, by bathing in the sea
or in other waters, by eating various foods. Pliny, Vergil, St Augustine
believed that mares could be fertilized by the wind. In northern India,
as in America, young women are careful not to Unger in the sunlight
while they are menstruating for fear of pregnancy. A Mongol princess
conceived through the operation of a hailstorm, and another by an
aurora borealis. We read similarly of the old inhabitants of Scotland
that a young girl, walking on an old battlefield, became pregnant when
a whirlwind threw some of the ashes in her private member.

In many stories of virgin births conception is caused by something
eaten; the virgin goddesses and princesses of China usually conceived
by eating a lotus flower; the Shang Dynasty was descended from a
princess who became pregnant through eating swallows’ eggs. The
Divine Mother in Japanese tradition conceived through eating cherries.
Hera conceived Hephaistos by eating a flower, and in the Finnish poem
of Kalevala, Mariatta conceives through eating a bilberry. In several
stories virgins conceive through eating bone-dust. Jacob, no doubt, was
conceived by Rebecca from her eating the fruit of the mandrake,

( Genesis xxx. 14-24). Throughout northern India coconuts are eaten to
cause conception, and Hindu women, Uke those in Tuscany and Portu-
gal, obtain consecrated food from priests with this object. The dew
from a mistletoe bough or an infusion of the plant caused animals and
women to become pregnant, according to the ancient Celts, from which
derives our own attitude to the mistletoe. The Ainu of Japan and the
natives of Mabuiag in the Torres Straits take a similar view.

Relation between Food and Offspring

Now, it is universally held that eating the flesh of an animal imparts its
qualities to the consumer. Courage is acquired from lions and the like.

THE TOTEM

259

swiftness from antelopes, ability to jump from kangaroos. Conversely,
eating deer makes men timid and eating the flesh of cattle makes them
slow. Eskimo children are given seals’ eyes to eat so that their eyes may
become bright and clear. The Greeks believed that to eat a nightingale
would cause insomnia.

Sometimes it is thought that the person will become completely
assimilated with the eaten. Thus in the Celebes a woman, after eating
snake’s flesh, turned into a snake.

The effect of food on pregnant women is particularly pronounced.
Indian women of the Gran Chaco do not eat mutton after marriage lest
their children be flat-nosed. The Ibibio of Southern Nigeria, like the
natives of Anatolia, think that if a mother eats snails her baby will slaver.
Similar prohibitions are found in every part of the world from South
Italy to Penrhyn Island. So numerous are these dangers that primitive
women are very restricted in diet. Gipsy women eat cocks or hens,
according to the desired sex of the child, and in southern Hungary they
eat slices of quince sprinkled with the blood of a strong man.

Most people believe that the fancies which pregnant women conceive
for certain food must be satisfied at all costs, and men have been known
to go fifty miles to obtain strange articles of diet, such as tadpoles or red
ants’ nests. The Hausa regard birth-marks as the result of leaving such
fancies unsatisfied, and in Italy they are actually called ‘voglie’ — that is,
‘fancies’.

The Importance of Food Animals

For primitive man the capture of animal prey must for long have been
far from an everyday occurrence. Even today it is often a red-letter day.
Savages are ravenously fond of meat, even where they are, from local
circumstances, mainly agriculturists. Among the Baganda, women are
said to have at times such craving for meat that they bite the ears of
their own children. The Wachanga of the Kilimanjaro region will travel
miles to have a drink of blood. The art of twenty thousand years ago
deals almost exclusively with animals. The animal on which his food
supply depended was the centre of man' s primitive interest ; it was his totem .
(The notion of an intimate relationship between a man and a natural
object, generally an animal, prevalent in many parts of the uncivilized
world, is known among the Ojibwa as a totem, from which the word has
been adopted.) Such traditions are also found in comparatively advanced
cultures, but, since many changes in the food supply must have occurred
in the course of migrations and altered conditions, we cannot expect to
find totemism still in its original primitive form. To interpret totemic
ideas is difficult. Though many totems consist of natural or even arti-
ficial objects, there is little doubt that they were originally animals or

260

THE MOTHERS

plants. The primitive identifies himself with his totem, believes himself
to be of the same species and endeavours to assimilate himself to it.
We find tribes whose totem is the wind or the sun, or even a tool or
weapon, but it is clear that a man could not identify himself perman-
ently with such things, and they must represent badges which are mere
survivals of the original totemic idea.

The totem is frequently tabu. It may not be eaten, killed or injured ;
sometimes it may not even be looked at, but this is comparatively rare,
and in Australia, as in North America, men kill the totem animals when
opportunity serves, though the Western Australian tribes will not kill
their totem ‘if they find it asleep’ and are said to ‘kill it reluctantly’. In
Chota Nagpur ‘the general attitude of the Oraon to his clan-totem is
that of a man to his equal, to his friend and ally’. Often the totem is
killed only after the formality of offering it an apology.

It is therefore difficult to believe that the totem originally had a
sacred character. As a rule, in primitive societies the awe with which a
sacred being is regarded tends to grow rather than to diminish. It is
therefore inconceivable that in societies which still possess a complete
totemic organization, this character could have wholly disappeared.
It is often suggested that where the totem is unceremoniously treated
totemic ideas are in a state of decadence, yet the whole organization and
many of the quasi-religious ideas of Australian tribes centre round the
totem, although it is clearly not tabu.

It is in Africa, where totemic ideas are overlaid with others and where
totemic organization is mostly in decay, that we commonly find rigorous
prohibitions and sentiments of awe attached to it. Further, the tabu on
eating the totem still has solemn force where every other trace of totem-
ism has disappeared; while in the more primitive totemic societies the
breach of food tabus is rarely regarded with horror. We must therefore
regard the tabu character as secondary.

Among many totemic peoples a breach of the tabu is not only permis-
sible but a duty, on certain occasions. Thus, the Edo of Nigeria must eat
a mouthful of the totem animal on the occasion of the annual festival
and at the funeral of any member of the totem clan. The food is merely
tasted; sometimes it is only taken in the mouth and spat out. Similarly
the King of the Banyoro, a pastoral people, fives normally on milk, but
must eat a small portion of veal at the chief ceremonies, not as a meal but
as a sacrifice to bring blessing on all the food of the land. Among the
Central Australian tribes many of the rites refer distinctly to such a
meal, although eating a totem does not today form part of the ceremony.
They rub the sacred ‘churingas’ on their stomachs, for they are sup-
posed to represent, or to contain, the essence of the totem. As they do so,
each man says ‘You have eaten much food’. These ceremonies take

THE TOTEM

26l

place at spots stated to be those where the ancestors of the clan cooked
and ate the totem for food. Although the known instances of the ritual
eating of a totem which is otherwise tabu are not numerous they are very
widespread, and they so closely resemble the ritual meals of higher
cultures that we must suspect that the latter had their roots in totemic
ideas.

Long before the ritual usages of totemic peoples were known,
Robertson Smith concluded that the ceremonial eating of the totem was
a feature of totemic belief and he illustrated his view by the description
left by St Nilus of the sacrificial eating of a consecrated camel among
certain tribes of Arabs on the Sinai peninsula. One of the rules was that
the whole beast must be consumed, hair, intestines and all, before the
sun rose. This rule is almost universal in primitive ritual meals, and was
observed by the ancient Hebrews when they partook of the Paschal
Lamb. We have noted it among the Australian Narrinyeri; it was ob-
served by the Algonkins. The Gallas of Abyssinia still practice rites
similar to those described by St Nilus; they say that the eating of
sacrificial flesh causes a spirit to enter their bodies. The Catholic mis-
sionary who reports this laughed when they told him; but when he saw
how scandalized they were at his impiety he became more serious and
he comments that ‘the conviction that a supernatural being enters into
them at the moment when they masticate the victim is akin to the
fundamental idea of the Holy Communion’. When these same Gallas, in
the seventeenth century, saw the Jesuits celebrating Mass, they were
convinced that the consecrated hosts were portions of the spinal marrow
of a hare.

The much-discussed question whether the eating of the totem is an
essential part of totemic practices appears to be misleading. If the totem
was originally the staple food of the tribe, to eat it would not be a cere-
mony but a necessity. Meal and ritual were originally one and the same.
The Algonkins eat the meal with just the same ritual observances as the
sacrificial communions just described, but this was simply the meal that
they ate when they returned from the chase with a supply of food. As a
Jesuit missionary comments : ‘The heathen do not partake of any meal
without making a sacrifice.’

Such meals are presided over by one of the older men of the tribe,
who open the proceedings with a solemn invocation ; and various rules
were observed, such as that no cutting instrument might be used. The
flesh of the animal had to be tom after the manner of primitive man. In
the laws of Leviticus, none of the bones of the animal might be broken;
the whole carcass must be consumed. These, in short, were ordinary
meals; the special significance which attached to them was as much a
question of the physiology of digestion as of the doctrine of trans-

262

THE MOTHERS

substantiation. In short, mystic and ritual meals were probably not
originally religious ceremonies with symbolic meanings but ordinary
meals. A ritual is a dramatic performance done in memory of the
significance which the act once possessed. Primitive man never merely
shoots an arrow, chops a piece of wood, lights a fire, but always accom-
panies such functional acts by words and actions of a magical character;
thus there is primitively no distinction between ritual and utilitarian
actions, between ‘religious’ and ‘secular’ acts. This is universally true;
among the ancient Greeks cooks and butchers were regarded as quasi-
priestly for, as Athcnacus knew, the butcher was originally a sacrificer
and the slaughtering of meat a priestly function.

In short, the totem was probably originally the food of the clan. In
several instances, we find it continuing to be the staple or favourite food
even when raised to a divine position. Thus the Huancas of Peru, who
worshipped the dog as a deity, ‘considered the flesh of a dog to be the
the most savoury meat’. ‘It may be supposed’, remarks Garcilasso de la
Vega, ‘that they worshipped the dog because they were fond of its flesh.’
Other Peruvians adored maize because it was their bread, and others the
fish, which they caught in great abundance ; while the Gollas worshipped
the white sheep, for they were owners of innumerable flocks. The God
of the Seri Indians is the great pelican, one of their chief sources of
animal food, although there is the germ of a tabu in the prohibition on
killing it during the breeding season. The Oraon clan regards its totemic
fish as tabu today, but tradition states that they once lived by eating it.
Some of the tribes of Western Australia were even more explicit.

Conception through eating the Totem

The transformation of the totem to a sacred food is logical enough for,
if the totem was originally the food of the tribe, it is truly from its sub-
stance that every member of the tribe was formed. This is in harmony
with the universal supposition that the child’s characteristics are
determined by the food which its mother eats. The Wogait of South
Australia say that when a man kills game he must give some to his wife,
believing that the food will cause her to conceive. Among the tribes of
the Cairn district of North Queensland acceptance of food from a man
by a woman is regarded, not merely as a marriage ceremony but as the
actual cause of conception. Elsewhere the man who kills an animal tells
its spirit, as it is dying, to go to a particular woman, from whom it is
thereupon reborn as a child. A native told Mr A. R. Brown that his
father had speared a small animal, now probably extinct, called a
‘bandary’, and he showed him a mark in his side where, as he said, he
had been speared by his father before being eaten by his mother. The
Eskimo and the Malays hold similar beliefs.

THE TOTEM

263

The presentation of food by a man to a woman is a customary wed-
ding rite in every part of the world. It is essential among North American
tribes, among the natives of Uganda and elsewhere. Even in the Upani-
shads we read: ‘Animals spring from seed, and the seed is the food.’

With the passage of time, the food supply changes; some animals
become extinct and their appearance is forgotten. When this has
occurred, a woman’s impregnation cannot be set down to an article of
food which she has not in fact eaten, and the theory must be adapted.

Where the totem animal has vanished entirely, dreams and impres-
sions must take its place as means of conception ; thus among Western
and Central Australian tribes conception is ascribed to the women
having been near certain spots where the spirits of unborn members of
a particular totem group are lying in wait, but we are told that these are
the places where the ancestors of the tribe actually ate the totem. Since
women who had actually eaten it would hardly ascribe the impregnation
to the action of the totem spirit at a distance, we must assume that this
represents an alternative way, developed when the practice of actually
eating the totem had ceased . 1

The Totem as Tribal Ancestor

The totem is also an expression of primitive social organization; the
unity of the clan is conceived as founded on the bond which the totem
constitutes. Most totemic people, asked to explain the significance of
their totem, say that it is their father or the ancestor of the clan. With
the most primitive totemic peoples the relation is not so much one of
historical ancestry as of identity of substance. An Australian native said,
pointing to a photograph which had been taken of him, ‘that one is the
same as me; so is a kangaroo’. The Bororo of Brazil, whose totem is a
red macaw, believe that they arc similar to it in appearance. They say :
‘We arc macaws.’ Many North American tribes think likewise.

As man comes to think himself descended from a totem ancestor rather
than identical with it, the totem tends to assume indefinitely human
attributes. Thus among the Alcheringa of Australia a man of the
Kangaroo totem may sometimes be spoken of as a man-kangaroo or as a
kangaroo-man, while the ancestor, by natural evolution, tends to assume
the character of a deity and a creator. Thus the Aborigines of Victoria
say that the world was created by the eagle-totem and the crow-totem.
The jay ‘was at that time a man’. The iguana ‘plays the part of an august
divine being’ at the initiation ceremonies of the Murring tribe.

J Sir James Frazer has set forth a theory of the origin of totemism which closely
resembles the one here put forward, but he thinks that it may cause conception
in a number of other ways besides being eaten, though, as he comments in many
of these instances, ‘this is totemism in decay’.

THE MOTHERS

264

As in Australia, so in North America. Among the Wyandots the sun
was created by the Little Turtle. The Manitu of the Delawares is an
elk; elsewhere he is a beaver, a bison, a squirrel, an otter or a dog. This
fusion between animals, men and gods is equally conspicuous among
the peoples of northern Asia. Thus among almost all Siberian tribes the
most important divinity is the Great Raven, and the Koryak declare
that in his time ‘there was no sharp distinction between men, animals
and other objects’.

Obviously such a totem tends to become sacred, and, from being the
ordinary article of food, it tends to become prohibited. By a common
subterfuge only the least attractive parts of the animal are eaten, a
device which was also found among the ancient Egyptians. Possibly some
of the split totems may have resulted from the segmentation of clans
originally having the same animal for their totem. However, no doubt
many totems never had any connection with the food supply of the clan,
for in every society new clans must of necessity come into being and
they must then assume a totem which is little more than an arbitrary
badge.

To recapitulate, then, we find a number of linked ideas:

(1) Many tabus refer to eating.

(2) A woman may be impregnated by eating.

(3) The child may be influenced sympathetically to resemble the
animal eaten.

(4) Men identify themselves with their totem animal.

(5) The totem may become the tribal ancestor.

(6) The totem animal is eaten, sometimes with apologies or precau-
tions.

This constellation of ideas seems explicable on the assumption that
totems were originally the animals on which the clan depended for sup-
port but which have since become rare, and rituals in which the flesh of
an ancestor-deity is eaten arise in this way.

Diffusion and Fundamental Conceptions of Totemism

Totemism, together with the exogamic totem-clan organization, exists
in North America, among some tribes of Central and South America, in
many parts of Africa, among several of the Dravidian races of India, and
is prominent in almost all of the Australian tribes. Conclusive indica-
tions of its former existence are found in Melanesia, New Guinea,
Micronesia and Indonesia. This makes it clear that we have to do with
one of the most fundamental conceptions of the human mind, and its
existence in Australia shows that such ideas date back to the Pleistocene
age. The inference appears irresistible that this conception was once

THE TOTEM

265

universal . 1 However, we are not justified in inferring, every time we find
a people worshipping a god in the form of an animal, that such people
once had an animal for their totem. It is improbable that bulls and
serpents, which are constantly associated with gods, were formerly
totems, for the symbolizm which has given rise to this association is
clearly traceable, although the process of making such an association
may derive from the traditional habit of associating animals with gods.

Furthermore, we come across conceptions identical with those of
totemic peoples among peoples who show no other traces of totemism ;
for instance, the Eskimo, than whom no race shows a more complete
absence of totemic organization . 2 Nevertheless, the conceptions we have
been discussing are as prominent in the ideas of the Eskimo as in the
most typical totemic society. The animals on which they live are not
regarded as prey, but with sentiments of sympathy and almost venera-
tion; the hunting of animals is subject to complex magical rites, designed
to conciliate them, so that they will allow themselves to be killed, and
this is done with apologies. After a seal, a walrus or a whale has been
killed mourning rites are observed with the same ritual as for humans.
The cooking and consumption of the meat are subject to ritual rules,
and certain parts of the animal must not be injured. The soul of the
slaughtered creature is supposed to reside in its bladder, and these
bladders are the most sacred of Eskimo objects. The food animals are
regarded as limbs of the Great Goddess Sedna, the moon, who is also
represented as a walrus ; at the same time, they are also represented as
human beings. The Eskimo also have a notion of conception similar to
that of the Australians ; a pregnant woman may only eat game which her
husband has killed, and should she eat any other animal food the hus-
band would not regard her child as his. The extent to which the
Eskimo retain or exhibit such ideas shows how fundamental they must
be, and perhaps the failure of their conceptions to develop into a social
tribal organization has helped to preserve their original character.

The primitive man’s belief that his strength is derived from his food,

1 The substitution of theriomorphic divinities as pseudo-totems for the ori-
ginal clan-totems is found in the lowest stages of culture. Thus the numerous
snake, bird and stone ‘totems’ of the Melanesian races of eastern New Guinea
are certainly not totems in the original sense. As Sir James Frazer says of the
divine beings, represented as a shark and a crocodile in Yam, ‘they seem to be on
the point of sloughing off their animal skins and developing into purely anthro-
pomorphic heroes or gods’.

Captain Cook directly observed the adoption of an animal as a totem on the
Island of Atiu where he presented some people with a dog. The clan then adopted
this animal as their totem and declared it was their ancestor.

2 Though some authorities declare that the Inuit have no totemic system, Mr
Nelson thought he noticed among the Eskimo of the Bering Straits some sug-
gestions of totemic institutions.

T.M — 9*

266

THE MOTHERS

and that his body and spirit were first formed from the food which his
mother ate, leads to magical practices designed to promote the food
supply, and when tribal groups are distinguished from each other, the
organization is founded upon such conceptions, but this organization
does not depend upon magical or religious ideas but is a corollary of
totemic ideas which are probably more general in distribution.

Priestly Clans

We can sec how, when totemism decays by the multiplication of totem
clans, a priesthood could emerge. The members of a totem can exert a
special influence over their totem animal : members of the Omaha bird
clan can stop birds eating the corn, and members of the Kansas wind
clan can create a breeze. When Mr Hollis’s porters in East Africa
were attacked by a swarm of ferocious bees and were compelled to
abandon their loads, a Nandi, belonging to the bee totem, volunteered
to quieten the bees, for, he said, ‘they belonged to him’. Naked, he
approached the spot and whistled the bees back into their hive without
being stung. If a man is injured by his totem animal, this is often taken
as a clear indication that he is not a true member of the totem. Ana-
logously, ceremonies calculated to promote fertility arc more effectual
when carried out by members of a totem clan in regard to their own
totem. Thus, such members enjoy a privilege belonging to them by birth.
The specialization of a particular totem-group in such matters benefits
the whole clan, and the clan whose totem has a general economic impor-
tance becomes a sort of magical society or, as it might be called, a col-
lege of priests. Such a group, therefore, comes to have magical powers
with regard to the ancestors of the clan, and so over all forces vital to the
clan’s welfare, such as the control of rain or luck in war. Such clans will
become sacred clans, royal clans; their power will be exercised on
behalf of the people to propitiate the god who represents the transformed
ancestral totem, and who can only be approached through their media-
tion. Their sacred character depends upon the fleshly kinship between
the priest and the god, and this identity with the god is reproduced by
rites of eating in many religions.

Primitive Tribal Solidarity

Such ideas have been the corner-stone of human social solidarity. The
members of the clan regard themselves as members of one body through
the community of food. In Arabic and in Hebrew the same word signifies
‘flesh’, ‘kindred’ or ‘clan’. A similar nomenclature is used in Tibet and
elsewhere. The word ‘ebussia’, for instance, is used by the Fanti of the
Gold Coast to connote either the totemic animal or the maternal family.
In the proto-human stage, the bond between members of a brood did

THE TOTEM

267

not rest upon a concept but upon a spontaneous sentiment, the irradia-
tion of the maternal instincts. A primitive man’s solidarity is represented
in his conceptions by the community of food. The Ojibwa Indians state
that they have totems in order that ‘they might never forget they were all
related to each other, and that in time of distress or war they were
bound to help each other’. It is a universal rule among totemic savages
that when travelling in strange territory a man enquires for persons
bearing his own totem, and is at once recognized by them as a brother.
When there is any possibility of disloyalty, solidarity is confirmed not by
rules and penalties but by exchanging blood and emphasizing the iden-
tity of substance. If the Arunta wish to prevent a stranger from revealing
their war plans, they force open his mouth and pour down blood of the
clansmen; after which it is inconceivable that he should give away their
plan. From this derives the conception of the relationship created
between host and guest by the partaking of common food, and Oriental
history abounds with instances in which the animosities of mortal foes
have been paralysed by their having taken food together. A robber who
gained access to the palace of the prince of Sagistan, happening to touch
some salt with his lips when he stumbled, had to withdraw empty-
handed.

As totemic ideas decayed, concepts derived from them took their
place — loyalties to divine kings, to a brotherhood, to the state, to the
throne — but they have proved poor substitutes. Sometimes we can see
among family groups in our own society the strange contrast of strong
loyalty within the group and complete callousness to those outside it.

Where the interests of one member are the interests of all, an injury
to one is an injury to all. There can be, as Mr Taplin says of South
Australians, ‘no personal property; all belong to the members collec-
tively and are the possessions of the clan. Each man cares for his neigh-
bour’s property because it is part of the collective wealth’. A fisherman
will call his friends to consult over the repairs to a canoe or to discuss
the marriage of his son. He is surprised that one should expect him to
act on his individual judgment; this would be dishonestly to ignore the
rights of others. Similarly the individual has no personal rights over
game or food which he may obtain, and the hunter may go short himself
that others should receive the recognized share. ‘Communism', says the
Rev. W. Ridley, ‘is another law of the aborigines. They hold the doctrine
of M. Proudhon, “La propriete e’est le vol”.’ The same surprising
ignorance of property has been noted among the Fuegians. ‘The perfect
equality among individuals composing Fuegian tribes is’, Darwin
thought, ‘fatal to any hope of their becoming civilized.’ A comparable
communism is practised among the Eskimo, the Veddahs, in Tahiti
and elsewhere. Of the North American Indians, Captain Carver says:

268

THE MOTHERS

‘The Indians in their common state are strangers to all distinction of
property, except in articles of domestic use.’ La Hontan notes that the
Indians ‘say that amongst us folks will rob, slander, betray, sell one
another for money . . . they think it strange that some should have more
goods than others . . . they never quarrel and fight among themselves,
nor steal from one another or speak ill of one another'. ‘What is extremely
surprising’, says Father Charlevoix, ‘is to see them treat one another
with a gentleness and consideration which one does not find among
common people in the most civilized nations. This, doubtless, arises in
part from the fact that the words “mine” and “thine”, which St
Chrysostom says extinguish in our hearts the fire of charity and kindle
that of greed, are unknown to these savages.’

The steadfast fidelity of such men to their fellow-clansmen, found also
among the Arabs, can be paralleled in Sparta.

Primitive human nature differed considerably from what we often
assume to be human nature in general. Primitives do not think in terms
of their ego and its interests, but in terms of the group-individual. Such
ideas are not the result of totemistic beliefs ; totemistic beliefs are the
result of such ideas. The individualism which is the alpha and omega of
the motives of modern man is not a primitive character but a product of
social evolution, which has developed mainly, if not solely, in relation
to social circumstances, and more especially to the growth of personal
property. Where primitive man has had a few years of contact with
Europeans a transformation occurs, and the chief cause is acquisition
of private property and his taking part in individual transactions. The
peasants of Europe closely resemble the savage in their traditions, but
differ from him in one respect — they are proprietors, and instead of
betraying sentiments of social solidarity they display narrow selfishness.
It is not the operation of individualistic instincts which has given rise
to the acquisition of personal property, but just the reverse.

CHAPTER 18

The Witch and the Priestess

Primitive Religion not Speculative

I t has been almost universally assumed that women have had little, if
any, pan in the development of religious systems, even though they have
often been their chief votaries. Such an idea should not be accepted as
self-evident. True, women are little disposed to construct speculative
systems. But it would be profoundly wrong to suppose that religious
ideas have grown out of intellectual systems ; rather the reverse. Primi-
tive man is strangely unconcerned about philosophical questions. The
following questions were put to a very intelligent Zulu : ‘Have you any
knowledge of the power by whom the world was made ? When you see
the sun rising and setting, and the trees growing, do you know who made
them and governs them ?’ The Zulu, after a thoughtful pause, replied :
‘No, we see them, but cannot tell how they come; we suppose they come
of themselves.’ 1

The Totem as Creator

The totem is by its very nature the tribe’s progenitor and creator; this
creative function is naturally extended to include the creation of the
habitat of the tribe, i.e. the world. The animal nature of ancestral totems
is, we have seen, but vaguely differentiated from their human nature.
The iguana, the turtle, the emu in Australia, the Great Hare, the White
Eagle in America, the Great Raven in Siberia, are all tribal Fathers,
supreme magicians, creators To the tribal ancestor are gradually
ascribed all the the functions tmceable to a superior agency controlling
the destinies of the tribe. He instituted the tribal customs of which no
one knows the origin, and is angry at any breach of them.

In more advanced stages the totem or mythic ancestor of the oldest
and most influential clan occupies a position superior to that of all other

1 Many similar statements have been cited to show that such savages have no
gods. This does not follow; it only proves that the savage’s conception of a god
differs from the theologian’s. Actually primitive man generally recognizes beings
who may readily be viewed as divine though in many respects most undivine;
when asked what beings he regards as possessing the divine attributes of the
theologian’s God, he has no hesitation in referring to them, and he recognizes his
questioner’s deity as being of the same order.

THE MOTHERS

270

tribal totems. He is not merely a clan ancestor, but a tribal ancestor, a
tribal god. He is not merely the source of a particular supply of food, but
a dispenser of subsistence, the controller of the tribe’s fortunes. The
dominant clan is the mediator between the tribe and the controller of its
destinies. The headman of the sacred clan is the representative of the
god, his ancestor; he is his earthly avatar, his incarnation, indistinguish-
able from the god himself. In Madagascar the King is simply known as
‘God on Earth’, and the creation of the world is ascribed to one of his
ancestors. The Queen of Angola, on being asked who made the world
and who fecundated the ground and ripened the fruits thereof, replied
without hesitation. ‘My ancestors’. ‘Then’, rejoined the Capuchin who
was catechising her, ‘does your Majesty enjoy the whole power of your
ancestors ?’ ‘Yes’, answered the Queen, ‘and much more, for over and
above what they had I am absolute mistress of the kingdom of Matamba.’

The divine king does not derive his status from any personal merit or
attributes, but from the fact of his descent from the god. With the
development, by ambitious conquerors, of royal power, the character of
the divine ancestor is enormously enhanced. To us the identification of
primitive kings with the supreme deity appears sacrilege or flattery.
But, as a matter of fact, it is not so much the royal personage who is
exalted by being assimilated to a god, as the primitive god who is
exalted by the attributes of his earthly representative. The rude tribal
god, who is in general treated with scanty reverence, first acquires
majesty when impersonated by a powerful monarch.

The ancestral god has a special function, closely related to his original
totemic character: rain-making. Whatever may be the source of a
people’s food supply, the most important factor determining its abun-
dance is the weather, and especially the rainfall. It is difficult for us to
realise fully the meaning of rain or drought to the primitive man. In
Australia a drought means misery and death to the blacks. In South
Africa, remarks Father Junod, ‘drought is equivalent to famine, and
famine to death’.

The rain from heaven has been the supreme determinant of the his-
tory of humanity. The great movements of pre-history, which have
determined the present distribution of human races, took place mainly,
if not solely, under the urge of the fatal drought. What is now the great
African desert was one of the first regions from which peoples were
driven northwards by the failure of the means of subsistence to people
Mediterranean shores and Northern Europe.

Thus it is not by mere poetic fancy that heaven is the abode of the
gods. The supreme gods of early religions not only dwell in heaven but
are the heavens or the heavenly bodies, thought of as controllers of the
seasons and of atmospheric conditions. Among the Tshi-speaking

THE WITCH AND THE PRIESTESS 27I

peoples the divine name ‘Nyamkum’ means ‘sky’ or ‘rain’. Among the
Makuas the same word means ‘sky’, ‘clouds’, or ‘God’. With the Basetos
of the Upper Nile the supreme god is simply ‘the rain-maker’. The same
nomenclature obtains in Asia. Among the Mongols the supreme god is
Tengri, ‘the sky’; in China, Ti, ‘the sky’; in Vedic India, Dyaeus, ‘the
sky’; in Persia, Ahura, ‘the azure sky’. In Greece he was Zeus, ‘the sky’,
‘the cloud-compeller’. Yahweh, the god of the ancient Hebrews, was a
rain-god ; ‘He shall come unto us as the rain’, says the prophet Hosea,
‘as the latter and former rain unto the earth.’

Thus it is that one of the chief, if not indeed the chief, functions of all
primitive priest-kings was the control of the weather, and more particu-
larly of the rainfall — as Frazer has pointed out. A king was primarily a
maker of rain, and originally was probably the headman of the clan
which was credited with the greatest power of wielding such control.
All African kings were primarily rain-makers. For instance, the function
of the King of Loango which most impressed the first missionaries was
his obligation to make rain. In Somaliland a chief is known as ‘Prince of
Rain’. Throughout the continent, in fact, ‘the chief was the great rain-
maker of the tribe’. Even the kings of ancient Egypt were rain-makers,
although Egypt, where it scarcely ever rains, would appear to be about
the last place for a rain-maker to set up. Pharaoh’s control of the waters
was naturally applied more often to regulating the river; and he caused
it to rise by casting into the Nile a written order to do so. Conversely,
the most important ceremony at the court of the King of Siam was that
at which the king, like Canute, issued through heralds a solemn order to
the River Meinam to retire. The Emperor of China also had a similar
office. Hindu doctrine teaches that Indra sends no rain upon a kingdom
which has lost its king. Ulysses explains to Penelope that: ‘Under a
virtuous prince the earth brings forth barley and wheat in abundance,
trees are loaded with fruit, ewes sit several times in succession, and the
sea is filled with fish. Of so great worth is a good leader.’

The control thus exercised by sacred kings over heaven for the benefit
of the people is thoroughly practical, and is exactly of the same kind as
that exercised by members of a totem-clan over their totem. The pur-
pose of primitive religion is eminently utilitarian. It is part and parcel of
the means employed to supply and control the necessaries of life, to
promote the prosperity of the tribe, and above all to provide its food.

‘The really important question’, as Robertson Smith remarked, ‘is not
what a god has power to do, but whether I can get him to do it for me.’
This is why those divinities which, judged by a theological criterion,
appear to correspond most closely to the conception of a god — the
creators and controllers of the universe — are in most primitive religions
not the objects of worship . Thus the Bahima, who worship a tribal

THE MOTHERS

272

ancestor, have also a god, Lugaba; but ‘they know very little about him;
he has no priests and so receives no sacrifices’. The supreme god of
Dahomey is ‘ignored rather than worshipped’. The same remarks apply
to all African populations. In West Africa generally ‘they regard the god
as the creator of man, plants, animals and the earth, and they hold that,
having made them, he takes no further interest in the affair’. ‘The god,
in the sense we use the word, is in essence the same thing in all Bantu
tribes I have met with’, says Miss Kingsley, ‘a non-interfering and
therefore negligible quantity.’

Similarly the American Indians ‘nowhere adored the god they knew’.
The tribes of Guiana, though they have the notion of a supreme being,
‘concern themselves little about him’. In Australia, Baiame, in whom
some enthusiasts thought they recognized a ‘supreme being’, is believed
by the Queensland tribes, says Mr Thorne, ‘to have gone away over the
ocean so long ago that our informant could give no idea of the lapse of
time, and never took further heed of the country or its inhabitants’. Of
Daramulum, another Australian ‘supreme being’, ‘there is no wor-
ship’.

The incongruity of divine beings who form no part of religion
‘becomes intelligible when it is borne in mind that primitive religion
refers to practical issues. The god in whom primitive man is interested
is not the sky-god, but the tribal ancestor who is also the supreme
magician, who can use his power to control the sky-god.

Since, then, primitive religion has little concern with philosophical
speculations, the natural inaptitude of women for such speculations is
irrelevant.

The Exclusion of Women from Religious Functions in the West

Nowadays, in Christian and other strongly patriarchal societies (e.g.
Brahmanical India or China), the notion of women exercising priestly
functions offends propriety. The suggestion that they should preach
and administer sacraments is regarded as an extravagance of feminism.
Such notions are comparatively recent.

The ancient world was full of priestesses. The Vestal priestesses were
one of the most ancient and sacred institutions of Roman cult. When
walking abroad, they were preceded by lictors bearing the insignia of
supreme command; any insult was punishable with death. In earlier
times they and other priestesses — the Regina Sacrorum, the Flaminicae
— played an even more important part in Italian cult. Ancient Italy
swarmed with priestly and prophetic women who often exercised
greater influence than the official priestesses. The sibyls of classical
tradition are the types of prophetic females, or shamanesses; the most
sacred shrines of Greece, such as those of Delphi and of Dodona, were

THE WITCH AND THE PRIESTESS 273

served by prophetic women. The priestess of Demeter, like the Vestals
at Roman spectacles, occupied a throne of honour at the Olympic
games. There can be little doubt that in these primitive cults, which
later became connected with Dionysos, priestly functions were exercised
exclusively by woman. In Aegean and Cretan religion, archaeological
evidence shows us priestesses discharging all religious functions.

‘As in Greece, so in Babylonia and Assyria’, says Professor Sayce,
‘women were inspired prophetesses of the god.’ In Assyrian inscriptions,
they are called ‘The Mothers’.

None but women were allowed to enter the Holy of Holies of Bel-
Marduk. In Carthage likewise women mediated between the Great
Goddess and the people. In ancient Egypt the Queen was high-priestess
of Ra. There were many orders of priestesses under the Old and the
Middle Kingdom ; and ‘at the time of the New Empire there was scarcely
a woman from the highest to the lowest who was not connected with the
service of the temples’.

Priestesses in Uncultured Societies

In primitive cultures, the part played by women in religious cult is
striking. In the state religion of Dahomey at least as many women as
men exercise priestly functions; priestesses undergo a three years’
course of initiation; they are called ‘Mothers’. Their person is inviolable,
and they enjoy great privileges. In most African kingdoms, such as
Ashanti, IJrua and Uganda, the temples of deceased kings are served by
colleges of priestesses and vestals. The most dreaded deity of Matabele-
land was served by a college of priestesses who were regarded as his
daughters; even King Lobengula had at times to yield to it. The numer-
ous female fetiches, or ‘Mother fetiches’, throughout Africa are served
exclusively by women. Some of them rise to positions of enormous
influence. In Loango, the priestess of Atida was called ‘The Mother of
God’. Throughout the Congo both sexes exercise magical and priestly
functions.

Among the Eskimo the shamans, or ‘angakut’, may be either men or
women. But Dr Rink maintained that formerly all magicians were
women. Today, in East Greenland, there are two classes of sorcerers, the
‘angakut’ proper and the ‘gilalik’, an inferior order. The latter are nearly
all women. Dr Rink believes that the male ‘angakut’ ousted them from
their former positions.

Among the North American Indians ‘medicine women’ were as
famous as medicine men, and on some occasions, such as the Com
Feast, they exercised almost unlimited authority. Their influence appears
to have been greater among the prairie and western tribes than among
the more advanced eastern nations. Among the tribes of California they

THE MOTHERS

274

are reported to have been particularly numerous. Among the Yurok
tribes of the Klamath River district the shamans ‘were almost all
women’. So among the Zuni, though today the rain-making ceremonies
are in the hands of priests, and although the priestly college consists of
men, at their head is a woman, the Priestess of Fertility. She can dismiss
any of the priests at a moment’s notice, without offering a reason. There
are several ‘secret societies’, but ‘there is only one person among the
Zuni who is a member of all the sacred societies and thus knows the
secrets of all, and that person is a woman’.

Competition often develops between the sexes for the possession of
the power derived from the exercise of magical and religious functions ;
accordingly they are frequently reserved to one or the other sex. But
such a monopoly is not characteristic of societies which have reached a
considerable degree of development under undisturbed matriarchal rule.
Thus, among the Khasis of Assam the priestesses perform all the rites
and sacrifices, but, as among the Pueblo Indians, the men are not
excluded, although ‘the male officiants are only the deputies’ of the
priestesses. The ultimate authority rests with the priestess, who is also
invariably the keeper of the sacred magical objects.

From a cursory perusal of the most accessible accounts of Central
America in the days of the European conquest, one would gather the
impression that, although there was a considerable sprinkling of shaman-
istic women, the wizard, or ‘paje’, was usually a man. Dr Brinton’s close
examination reveals a very different state of things. The ‘pajes’ were
members of a closely organized and widely spread association which has
been termed Nagualism . 1 ‘A remarkable feature of this mysterious
organization’, says Dr Brinton, ‘was the exalted position it assigned to
women. Not only were they admitted to the most esoteric degrees, but
in repeated instances they occupied the very highest posts in the organ-
ization.’ Pascual de Andagoya asserts from his own knowledge that some
of the female adepts had attained the rare and peculiar power of being
in two places at once. Spanish writings of the sixteenth and seventeenth
centuries confirm the dread in which they were held. ‘In the sacra-
ments of Nagualism, woman was the primate and hierophant’, Brinton
declares.

In Guatemala the supreme ministrant of the gods was a priestess, and
it was to her that the warriors applied to ensure victory. ‘In many native
American legends, as in others from the old world, some powerful
enchantress is remembered as the founder of the state. Such among the
Aztecs was the sorceress who built the city of Mallinalco, famed even

1 The notion, common also in other parts of the world, that the shaman can
transform himself into an animal, and that his power is somehow bound up
with a spiritual double, or ‘nagual’, dwelling in the animal.

THE WITCH AND THE PRIESTESS

275

after the conquest for the skill of its magicians, who claimed descent
from her. Such in Honduras was Coamizagual, Queen of Ccrqui, versed
in occult science, who died not, but at the close of her earthly career rose
to heaven.’

The position of sorceresses in South America would appear to have
been similar. Among all the tribes of the Amazon ‘old women are the
interpreters of the gods’. In Patagonia ‘the old women, witches, pro-
phetesses or divineresses are the chief ministers of their cult’.

Throughout the Indonesian archipelago the primitive aboriginal cults,
where they have not been supplanted by Islam and other foreign religions,
are predominantly and often exclusively served by women. Thus among
the Bataks of Sumatra the shamans may be either men or women; but
female shamans are far more common, and in several districts there are
none but female shamans.

In numerous other Pacific islands, such as Timor and the southern
Moluccas, the shamans are predominantly women.

Among the Dayak tribes of Borneo religious functions are almost
exclusively exercised by women; the shamans ‘are for the most part
women, seldom men’. The same is true of the southern Celebes, where
they are knowm as ‘imitation women’. All the deities or spirits from
whom sorcerers, whether male or female, derive their power are spoken
of as their ‘grandmother’. According to the tradition of the Land Day-
aks, the magic art was first imparted by Tuppa, the women’s deity,
to a woman, and was taught by her to her successors. ‘It seems to me
more than likely’, says Miss M. Morris, ‘that manangism [shamanism]
was originally a profession of women, and that men were gradually
admitted into it, at first only by becoming as much like women as
possible.’

We may here note that any attempt to draw a sharp line between the
private magical practices of the shamaness and the more dignified office
of the official priestess proves futile in practice. Thus in Indonesia,
where the people themselves draw a distinction between the shamanesses
and the priestesses, both Dr Wilken and Dr Kruijt, the highest authori-
ties on the religious usages of the region, are compelled to admit that
there is no psychological distinction between the two. Both priestesses
and shamanesses are in exactly the same condition of spiritual posses-
sion, and the magical procedure ritually observed in the public cult is
identical with the practice of the individual witch or shamaness. The
distinction between the two is not religious or psychological, but social
and official.

Among the Polynesian races, who originally came from the Indone-
sian region, power has become concentrated in the hands of chiefs and
the aristocratic classes. To this is due both the decay of popular religious

THE MOTHERS

276

cult — the priestly offices being for the most part under the jurisdiction
of the chiefs and ruling classes — and the masculine exclusiveness, if not
the patriarchal character, of Polynesian institutions. Women were, as a
rule, excluded in Polynesia from the sacred rites of men, so it is all the
more remarkable that official priestesses, occupied a high position in
many Polynesian islands. Thus there were priestesses in Rotuma,
Tonga, Samoa, Paumotu, Uvea and Savage Island.

‘Nearly all writers on Siberia agree that the position of the female
shaman in modern days is sometimes even more important than that
occupied by the male. . . . Among the Palaeo- Siberians, women receive
the gift of shamanising more often than men.’ According to the tradi-
tions of the Yakuts, there were formerly no male shamans, or priests, but
all magic functions were exercised by women. This was still the case
until recent times among the Kamchadals and the neighbouring popula-
tions. The familiar spirit from whom every practitioner of the magic
arts is supposed to derive his or her power is spoken of among the
Yakuts as his ‘Mother’. In the languages of the Yakuts, Altains,Torgut,
Kidan, Mongols, Kirghis, Buryat, the term for shamaness is the same;
while quite different words are used to denote male shamans. From this
Troshchanski infers that, before the separation of these races, all
practitioners of shamanism were women and that male shamans only
appeared subsequently. Male shamans among the Yakuts wear long hair
and dress as women, whether they are wearing their ordinary dress or
their ceremonial costume, and two iron circles on their apparel repre-
sent a woman’s breasts.

Male Priests dressed as and impersonating Women

The adoption of female dress by male shamans and priests is a world-
wide phenomenon. It was prevalent among the North American Indians;
invariable throughout Indonesia. In Tahiti and the Marquesas the
priests of the Areois stained their skin a light or more feminine hue and
affected the manners of women. When Zulu chiefs perform rain-making
ceremonies they put on a woman’s petticoat. Among the ancient Germans
male priests dressed as women. In Babylon the priests of Ishtar and the
Syrian goddess wore female attire. So likewise did the Korybantes, the
Dactyloi, the Kouretes and the priests of Artemis at Ephesos. The
priests of Herakles at Kos dressed as women when they offered sacrifice,
as did male officiants in the festivals of Dionysos. The male assistants
in cult scenes from Minoan Crete are represented wearing women’s
clothes. All priestly robes, skirts, aprons, sottanas are indeed everywhere
of an essentially feminine character.

On the other hand, instances of a woman dressing as a man when
exercising priestly functions are altogether exceptional, although women

THE WITCH AND THE PRIESTESS 2JJ

dress as men when exercising any prescriptively male occupation, such
as soldiering or hunting . 1

In primitive societies ‘wearing the clothes that have been used by
another transfers to another wearer the qualities of the former one’.
Similarly it is a universal principle that the distinctive dress of each sex
implies that the person wearing it is engaged in the occupations which
are peculiar to that sex. Thus for instance, among the tribes of California
every warrior, when too old to take part in active warfare, assumes
female attire, and thenceforward helps the women with their duties.

It is frequently stated by those reporting that among a given people
certain men dressed in woman’s clothes and followed women’s occupa-
tions, that such men served for the indulgence of unnatural vices. But
it may be said positively, at least so far as regards the North American
tribes, that no grounds exist for this inference.

In primitive societies the assimilating of men exercising priestly
functions to women goes much farther than the assumption of feminine
apparel. In the Pelew Islands, for instance, ‘it often happens that a
female deity chooses for her priest a young man, who is thenceforth
regarded and treated in every respect as a woman. He assumes female
dress, and wears a piece of gold round his neck, and he also frequently
takes up the cultivation of a patch of taro.’ In Cyprus, at the festival of
Ariadne, the imitation was carried even farther — for one of the officiating
priests lay in bed and imitated the groans of a woman in labour. Among
the Yakut it was actually believed that male shamans were capable of
bearing children. In California the male shamans assume female attire
because this is regarded as bestowing greater power, and the same is true
of the Chukchi. In short, this universal practice does not seem open to
any other interpretation than that magic was orginally regarded as
essentially a woman’s function.

It is a curious fact that smiths are widely held to possess magic
powers. Yakut traditions connect the appearance of male shamans
amongst them with the introduction of iron. The first male shamans,
they say, were smiths. The profession of smith is hereditary among
Siberian tribes, and the Yakuts consider that at the ninth generation a
smith becomes a wizard. Among the Buryat the spirits from whom men
derive magical powers are called ‘smiths’, and are thought to have
taught men the ironworkers’ art also. A proverb of the Kolyma district
affirms that ‘the blacksmith and the shaman are of one nest’. Among the
Mongols the same word denotes a male shaman and a smith. Among

1 The women of the Mawungu secret society among the Pangwc dress as men
at their festivals. The ‘medium* of the god Mukasa in Uganda adjusts her clothes
in male fashion when she is acting as the mouthpiece of one god; but at other
times she remains purely feminine, and is, indeed, regarded as the wife of the god.

THE MOTHERS

278

the Romans ‘faber 5 , smith, connoted ‘magician 5 . In Russian popular
tales smiths act as assistants to witches. The Kayan Dayaks believe
smiths to be possessed by spirits, and that their skill is due to this.
Similar estimates are general in Africa. Thus among the Fans the village
blacksmith is the priest and sacred headman of the community. Tribes
ignorant of the art of metallurgy regard smiths with such awe that if they
obtain possession of a smith's bellows they place it in their fetich-house
and address their prayers to it. The Arabs and Berbers ban all smiths
from society. Dr Schneider is probably right in supposing that this
‘indicates that the art of the smith is regarded as a branch of witchcraft,
and those who practise it as the possessors of magical powers 5 . In ancient
Egypt the priests of Horus were known as smiths. In Asia wizards were
the particular disciples of Tubal-khan, the smith. In Ireland St Patrick
pronounced against ‘the spells of women, of smiths and magicians 5 .
Such views accord with the belief (widespread in Europe) that smiths
are the only men who share the magical powers peculiar to women.

Priestesses among the Peoples of Northern Europe

In the druids we appear to have a purely male priesthood. Their origin
is disputed. Some authorities hold that they were always the priests of the
Celtic races, others that they are of late origin. It is improbable that a
highly organized religious corporation could have existed from the first
in all Celtic countries. Caesar, who enjoyed the friendship of the druid
Divitiacus, tells us explicitly that ‘the discipline [of druidism] is held to
have had its origin in Britain, and to have been transported thence into
Gaul . . . 5 . The facts appear to confirm this. The druids were certainly
much more firmly established in Ireland and western Britain than in
Gaul. They are said to have ‘tamed the people as wild beasts are tamed 5 ;
this would hardly have been needed if they had been their spiritual
rulers from the beginning.

That druidism was an importation at the cost of some earlier system is
suggested by other facts. When Hannibal passed through Gaul it was
agreed that all damages which might be caused by the transit of his huge
armaments should be assessed by a council of women. It is scarcely
conceivable that, had there been a powerful theocracy of druids at the
time — and we hear nothing of them in this connection — they would have
kept out of the whole business. Caesar, it is true, does not mention any
priestesses among the Gauls, though he states that the troops of Ario-
vistus were forbidden by their wise women to fight before the new moon;
and Gallic women attached to the armies offered sacrifices and prophe-
sied. (Some have spoken of ‘druidesses 5 , but there appears to be no
evidence of any women being connected with the organization of the
druids.)

THE WITCH AND THE PRIESTESS 279

Moreover, there is abundant evidence of the existence of priestesses,
who appear to have been thrust aside by the druidical invasion in Gaul.
Thus Pomponius Mela, in his brief reference to Gaul, observes: ‘Sena
in the British Sea is famous for its oracle of a Gaulish god, whose
priestesses, living in the holiness of perpetual virginity, are said to be
nine in number. They call them Gallizenae, and they believe them to be
endowed with extraordinary powers, so that they are able to rouse the
sea and the wind by their incantations, and to turn themselves into
whatsoever animal form they choose; they can cure diseases which are in-
curable to anyone else; they know the future and vaticinate .’ 1 Exceptional
interest attaches to the account, for it enables us to detect tradition
evolving. Sena is undoubtedly the island of Sein off the coast of Ar-
morica, far-famed in Celtic legend as the site of the grave of Merlin.
Now, in these (later) legends Merlin left the court of King Arthur
accompanied by ‘nine bards’. In other versions of the old druidical tales
Merlin fell into the toils of the ‘Lady of the Lake’, who became so
proficient under his instruction that one day, when he was asleep, she
cast a spell over him, making him a perpetual prisoner in a castle or,
according to other accounts, an oak-tree, whence he delivered oracles.
Thus the oracular divinity of the priestesses of Scin appears to become
a druid; but the priestesses of Sein do not appear to have had, in fact,
any connection with druidical institutions.

This is not the only institution of the kind. A little farther south, on
an island at the mouth of the Loire, near the present city of Nantes,
there was, says Strabo, another great shrine served by a college of
priestesses. They were said to celebrate ‘the mysteries of Dionysos’, and
their cult is described as orgiastic. So little had the cult to do with
druidism that no man was permitted to approach the shrine. Dionysius
Periegetcs tells us that in some of the Channel Islands the rites of
Bacchus were performed by women crowned with leaves, who danced
and made an even greater shouting than the Thracian bacchantes.

Similar evidence meets us in Ireland. In County Kildare is a monas-
tery of nuns dedicated to the service of St Brigit. One of their chief
duties was to tend, like the Roman Vestals, a perpetual fire. As Canon
McCulloch remarks : ‘The nuns who guarded the sacred fire of Kildare
had evidently succeeded the virgin guardians of a sacred fire, the pries-
tesses of a cult which was tabu to men.’ This is likely enough, for Brigit
was the chief Celtic goddess. It therefore appears improbable that her
cult elsewhere, in Britain and in Gaul, was originally served by male
druids. There are records of other shrines in Ireland where vestal fires
were maintained. If cults existed in Ireland, the stronghold of druidism,

1 ‘Gallizenae* is obviously a MS. error for ‘Gallae Senae’, or some such
expression.

28 o

THE MOTHERS

in which druids took no part, it can cause little surprise that in other
Celtic regions, such as western Britain, druids are scarcely heard of at
all. The famous queen of the Iceni, Boadicea or Boudicca, was the high-
priestess of the cult of the goddess Andaste. Pliny, again, refers to the
rites of the British women who, probably in ceremonies of agricultural
magic, dance naked, painted with woad.

It would thus appear that the exclusively masculine magical organiza-
tion or ‘secret society’ of the druids was superimposed at a comparatively
late date upon other cults in which the priestly functions were exercised
chiefly, perhaps exclusively, by women. ‘There is evidence’, Canon
McCulloch writes, ‘that they [the druids] had ousted women as the
earlier magic-wielding persons. ... In Irish texts women as magicians
performing all the magical rites ascribed to druids are much in evidence.
But their magic was, so to speak, not official.’

Among the Teutonic and Nordic barbarians, on the other hand, no
male priesthood ever became established. In nearly every instance where
a priestly personage is mentioned it is a woman. The influence and
power exercised by sacred women and prophetesses are frequently
referred to, and several were actually worshipped as divine beings during
their lifetime. Strabo reports : ‘They say that the Cimbri had the follow-
ing custom; their women, who travelled with them, were accompanied
by sacred priestesses, grey-haired, white-robed, with a linen scarf
buckled over their shoulder and a girdle of brass, and walking bare-
footed. These priestesses, with a sword in their hand, met the prisoners
of war when they were brought to the camp; and, having crowned
them, they led them to a brass basin as large as thirty amphorae. They
had a ladder, which the priestess mounted, and, standing over the basin
she cut the throat of each prisoner as he was handed up to her. With the
blood that gushed into the basin they made a prophecy.* One such
cauldron has actually been discovered at Gendestrup in Jutland; the
sacrifice is depicted on it; among the deities depicted on it is a moon-
goddess . 1

Among none of the Nordic races had the practice of the magic art
been taken over by men at the time when Christianity was introduced.
‘Our earliest antiquities’, says Jacob Grimm, ‘impute it pre-eminently

1 Cauldrons similar to that of Jutland played, as we shall see, a very important
part in Celtic religious ideas, and a number of epic myths relate the adventures of
heroes who undertook to obtain the priceless gift of a sacred cauldron, or
‘cauldron of regeneration’, as they were sometimes called, from some divine
woman. Those ‘cauldrons of regeneration’, which barbaric heroes coveted, were
transformed in Christian times into the vessel of the Holy Grail; and thus the
gruesome rites of the Cimbrian priestesses and the unholy mysteries of Celtic
witches are intimately connected with the sublimest heights of mediaeval
Christian mysticism.

THE WITCH AND THE PRIESTESS 2%1

to women.’ Thus among our own ancestors in Western and Northern
Europe, as elsewhere, it would appear that formerly religious and
priestly functions belonged originally to the women.

Traditions of Transfer of Religious Function from Women to Men

Further evidence of our thesis is afforded by the fact that in many parts
of the world we find traditions which suggest that religious functions,
now exercised by men, were formerly the women’s prerogative.

In Australia, for instance, as we have seen, the women formerly
played a much more important part in religious functions than now,
and the traditions suggest that formerly women were not excluded, as
they are now. They ascribe to women the magic powers today exercised
exclusively by the men. There is definite evidence that the tooth-
knocking ceremony, the mark of initiation, was performed on the
women as well as the men, among some tribes, as recently as the first
discovery of Australia by Europeans. Moreover, ‘in tradition after
tradition, we have accounts set out in great detail of how particular
women of the Alcheringa carried sacred Nurtunja just as the men did,
and how they performed sacred ceremonies exactly as the men did’.
The Queensland natives explain that women have a natural aptitude for
magic, which is precisely why they are debarred from practising it.

In several parts of the world traditions refer explicitly to the former
exercise of magical and religious functions by the women. Thus the
Fuegians still, like the Australians, use their religious ceremonies as a
means of frightening the women and keeping them in subjection, but
formerly it was the women who dressed up as ghosts and frightened the
men. Tradition says the change was brought about by a revolution in
which most of the women were massacred. The great religious yearly
festival of the Fuegians is supposed to commemorate the event.

The South American tribes of the Upper Amazon and Rio Negro
basin, particularly those dwelling near the Rio Uaupes, have similar
ceremonies. The traditions of the Uaupes Indians, however, state that
these ceremonies were formerly performed by women, and that their
rites were instituted by a ‘council of women’.

We come upon exactly similar traditions in Africa. In East Africa,
among the Wanika, a terrorizing cult exists identical in social use with
those just noted. In one district, however, at Rabbai Mpia, the situation
is reversed; there is a society which is confined to the women, who
simulate ghosts to terrify the men. Such women’s religious associations,
(or, as they are somewhat inappropriately called, ‘secret societies’) are
common in West Africa. There is reason to believe that almost every
woman belongs to such a society. What is known of their rules suggests
that women were formerly leaders of society and of religious cult. One of

282

THE MOTHERS

the most famous and powerful is the ‘Njembe’, to which every Mpon-
gwe woman is expected to belong. Initiation was formerly regarded as a
sort of religious duty and as a sanctifying act. There are various degrees
of initiation, and there is a supreme head, or Mother.

Europeans who have endeavoured to obtain a glimpse of the rites of
the ‘Njembe’ have barely escaped with their lives. Hence little is known
about them except that the meetings are held in secluded glades in the
forest and involve a sacred fire. The women (who are normally modest)
strip naked and dance until they are exhausted; phallic symbols and
fescennine songs are part of the ritual, which is also associated with
serpents, and each women must catch and carry one of the small snakes
which live among the mangrove roots. In Liberia, ‘If the tribe decides
to go to war, the declaration of war is not complete until it has been
referred to the women and they have approved of it.’

In the case of the Attonga sisterhood of Sierra Leone, ‘If a man hap-
pens, through ignorance or inadvertency, to enter an Attonga house, he
is made one of the society, though contrary to his inclination.’ He is
henceforth regarded as a woman, and when he dies he cannot be buried
in the men’s graveyard.

Special interests attaches to those societies admitting both men and
women. In the southern Sudan, the Bir society is open to both sexes;
but the women perform the essential ritual of maintaining the sacred
fire. So is the secret society of the Butwa, a timid tribe of fisherfolk
dwelling round Lake Bangwenlu, It is governed by five officials of each
sex, but the women play the chief part. The female hierophants are
called ‘the mothers of the Butwa mysteries’. Butwa women compel their
husbands to join the society. In the Purrah society of Sierra Leone,
whenever a man is admitted, he must be introduced and accompanied
by some female relative who is initiated at the same time. The head of
the society is a woman. The neophyte swears allegiance to the high-
priestess, saying, ‘You are my mother; I cannot betray you.’ Tradition
relates that the society was formerly exclusively female.

What is more, some of the most influential religious societies of West
Africa, from which women are at the present day excluded on pain of
death, were, we are told, instituted by women and were once exclusively
women’s societies. One such is the Egbo, which flourished among the
Efik and Ekoi people. On inquiring why their ceremonies sometimes
involve an old woman, Mr Amaury Talbot was told that the Egbo was
originally a women’s society, and that the men had learned their rites
and then forbidden the women to participate. At the New Year yam
festival, the members of another man’s society assemble and perform
rites, which include songs by a member chosen for his sweet voice. He is
dressed in women’s clothes and is called ‘The Mother of Ekong’. This

THE WITCH AND THE PRIESTESS 283

society is also declared to have been formerly an exclusive women’s
society.

Associations similar to the ‘secret societies’ of Africa are found in
most parts of Melanesia. In Duke of York Island the most important
secret society is the Dukduk, from which the women are rigidly ex-
cluded. Yet we are told that ‘the first “dukduk” was found by a woman
at Birara, New Britain, floating on four coconuts. She dressed it, and
soon exhibited it, and got lots of money. The men, however, got
jealous, and said that women were not tall enough for it, and so they
bought it and forbade women to go near it ever afterwards.’ As in the
West African societies, the head of the Dukduk societies, who is called
‘the Old Woman’, is a man dressed in female attire, and is often spoken
of as ‘she’. In British New Guinea, again, the initiation ceremonies of
young men, which constitute the chief religious observances, are said
by the Masuigam to have been instituted by a woman.

We thus find identical traditions in such widely different parts of the
uncivilized world as Australia, Ticrra del Fuego, the Amazon Valley,
West Africa, Melanesia and New Guinea, to the effect that the rites
from which women arc at the present day excluded were once either
instituted by the women, or that they took the leading part in them.
These traditions reinforce the conclusion which has been reached
independently in regard to Indonesia, America, Northern Asia and
Northern Europe, that magical practices and primitive priestly func-
tions formerly belonged to the exclusive sphere of women and that they
were taken up by men at a comparatively late epoch.

Powers of Witchcraft ascribed to Women

In its primitive phases religion is indistinguishable from magic. All
religion began as practical wonder-working. The distinction between
magic and withcraft refers not to the means employed nor the nature of
the powers used, but to the purpose to which they are put.

It is therefore relevant to our thesis that the power of witchcraft is
universally regarded as pertaining specifically to women. The witch is a
woman; the wizard is a male imitation. In primitive thought every
woman is credited with the possession of magic powers, wherever such
powers are believed in. Thus the Chukchi declare that ‘Woman is by
nature a shaman’. In New Guinea ‘the people will have it that all evil
spirits are female. They are all women or enter women, giving them
terrible power.’

In many different places tradition holds that women first taught the
magic arts. In Sierra Leone, although (as in other parts of Africa)
witches may be male or female, yet the power of witchcraft, which is
considered to be inherited, can be derived only from the individual’s

THE MOTHERS

284

mother. In India* according to Molwa, every woman was suspected of
being a witch, and the same test as in Europe was applied of ducking
her in a pond to see if she would float. In ancient Babylon the same
notions were held.

Since it is a common notion that such power is counteracted by child-
bearing, the power of witchcraft belongs particularly to old women or
to young unmarried women. Young and beautiful women are often
‘bewitching.’ This notion is explained by Dalla Porta, who traces
‘falling in love’ to the magical properties of menstrual blood. ‘If a person
is ensnared with the desire of a fair and beautiful woman, although he
be caught at a distance, yet he taketh the poyson in at his eyes, and the
image of her beauty settleth in the heart of this lover, kindleth a flame
there, which will never cease to torment him. For the soft blood of the
beloved, being strayed thither, maketh continual representations of her :
she is present there in her own blood. But it cannot settle or rest there,
for it continually endeavoureth to flye homeward, as the blood of a
wounded person spirts out on him that giveth the blow. But if it be a
fascination of Envy or Malice, that hath infected any person, it is very
dangerous, and is found most often in old women. And you will find
more women than men witches.’

A seventeenth-century Italian bishop devotes a chapter to discussing
‘Why witchcraft appertains to women and not to men’. ‘For one wizard
or necromancer that one may sec [he notes] one finds ten thousand
women.’

Modem anthropologists have generally sought to account for this
predominance by reference to the nature of women. Women, it is said,
suffer more from nervous instability than men; they are subject to
hysteria and temporary delusions. Such nervous disturbances are similar
to, or identical with, the state of ‘possession’ in which primitive pries-
tesses and shamanesses become the medium of supernatural powers.

I regard this explanation as inadequate. Amongst primitive races
nervous and hysterical phenomena are found almost as commonly in
men as in women. ‘Arctic hysteria’, which in Northern Asia is regarded
as closely connected with shamanistic powers, is reported to be common
among both sexes. The analogous phenomena noted in Indonesia
appear to relate almost exclusively to men. Again, men of Australian
tribes commonly have attacks of hysteria, during which they yell, foam
at the mouth and dance till exhausted; nothing is said about the women.
In any case, it is not necessary to be in a hysterical fit to work witchcraft.

It seems unnecessary to call on such theories. Though women are
commonly supposed to exercise power by virtue of their being witches
it seems equally probable that they were originally regarded as witches
because they exercised power. Where, as among the Seri Indians, the

THE WITCH AND THE PRIESTESS 285

elder women were genealogically and socially heads of the groups,
mediation between their children and supernatural powers would
naturally fall to their lot. The primitive mother is, by virtue of natural
position and function, the wielder of domestic magic. Among the
Chukchi it is the mistress of the house who applies the sacred paint to all
members of the family; she has charge of the sacred objects, and per-
forms all the religious functions connected with the household. ‘Conse-
quently the women are more expert than the men in the details of the
ceremonial’, and ‘the incantations and spells which are connected with
household charms are better known by the women’. The same is true of
the Eskimo. Among the Patagonians the magic or religious functions of
the household are the exclusive concern of the mother of the family.
Among the ancient Germans, Caesar tells us, ‘it was the custom that the
mothers of the families should declare unto them what they should do
through divinations and vaticinations’. Thus, if our conclusions about
the status of women in primitive society are correct the universality of
the attribution to them of magic powers is self-explanatory.

The source from which magic powers are regarded as being primarily
derived is, we shall see, closely connected with the physiological func-
tions of women; the magic faculties which it imparts to them are, ac-
cording to primitive conceptions, as much a part of their natural con-
stitution as are their reproductive functions. This dread-inspiring power
was primitive woman’s natural means of enforcing her authority — her
substitute for physical force. Her power was that of pronouncing curses,
of casting spells. It was by that power, I believe, that sexual tabus were
originally imposed upon mankind.

The diabolic nature ascribed to women, not only by the Christian
Fathers but by all humanity, is rather the expression of the dread with
which women were originally regarded than the cause of that dread.
Inevitably, men revolted against that source of terror. On the other hand,
it is quite inconceivable that female witchcraft should have arisen as an
imitation of practices invented by male magicians, and then have come
to be universally regarded as a faculty pertaining to the very nature of
women.

Witchcraft and Religious Magic

It may seem paradoxical to say that the priest or priestess acts by virtue
of the very powers upon which the witch or wizard depends. But, in
fact, the distinction between good and bad magic depends on the use
which is made of such powers in any given instance. The priest who
bewitches and destroys the tribe’s enemies appears maleficent to those
same enemies.

Thus ‘among the Matabele it is well understood that there were two

286

THE MOTHERS

kinds of witchcraft. One was practised by the witch doctors and the
king, such as, for instance, the “making of medicine” to bring rain, of
the ceremonies carried out by the witch-doctors to appease the spirits of
ancestors. The other witchcraft was supposed to consist of evil practices
pursued to cause sickness and death.’ The only difference is that the
user of white magic practises it openly, whereas the practiser of black
magic does so secretly. But ‘the same “medicines”, the same dances, the
same enchantments are used in both’.

Among the Eskimo, as among every other uncultured people, the
practice of witchcraft is abominated; but the magic arts which consti-
tute the ritual of actual religion arc in every respect identical with the
magic resorted to for private, maleficent purposes. The Samoyeds and
Lapps make no distinction between ‘white’ and ‘black’ shamans, but
every shaman may ‘serve both for good and bad ends as occasion
arises’.

Nor can we easily distinguish the witch from the priestess by referring
to the deities from whom they derive their power, as our forefathers
attempted to. Frequently the distinction between good and bad deities
is not made. The same deity can bestow or withhold fertility, can send
disease or withdraw it. Speaking of the divine beings of the Siouan
tribes of North America, Mr Pond remarks that ‘evidence is wanting to
show that the people divide these “Take-Wakan” into classes of good
and evil; they are simply “wakan”.’ Among all the people of Northern
Asia, says Mr Staling, ‘there is in the spiritual world of shamanism no
absolute contrast between good and evil’.

However, even where beneficent and maleficent deities are recognized,
it is more frequently the maleficent one who is the main, or sole, object
of religious cult. The natives of Kadiak ‘believe in a good and a bad
spirit; they worship the latter because they are afraid of his ill-will, and
do not sacrifice to the former because he will cause no harm to anyone’ ;
similarly with the Patagonians. The Yezidis of Armenia, at Easter,
sacrifice one sheep to Jesus Christ and thirty sheep to the Devil,
‘because’, they say, ‘he is more difficult to propitiate’.

Primitive supernatural beings are essentially maleficent. In primitive
society, power is synonymous with power to harm. The Santal of Bengal
‘cannot understand how a being can be more powerful than himself
without wishing to harm him’. The natives of South Africa represent
their gods as ‘mischievous, delighting to torment them in various ways’.
The Hottentots ‘have a vague notion of a benevolent spirit, but have a
much clearer notion about an evil spirit whom they fear, believing him
to be the occasion of sickness, death, thunder and every calamity that
befalls them’. Sir Richard Burton spoke to some African Essas about
God. They eagerly asked where he might be found, in order that they

THE WITCH AND THE PRIESTESS 287

might kill him, for ‘Who but he’, they said, ‘lays waste our homes, and
kills our wives and catde ?’

While maleficent magic is universally abhorred, beneficent witches
are venerated. In Russia they ‘stood high in popular estimation’, and
were thought in some respects more sacred than the Christian priests.
Even in Scotland, where the persecution of witches was once so severe,
witches have been held in high honour. In some districts fishermen
would not put to sea until a witch had performed incantations to secure
fair weather and a good haul. Conversely, priests are often called upon
to perform deeds of ‘black’ magic. Thus, on the Gold Coast they are
often asked to procure the death of persons who have offended the
applicants. The god who is thus ‘induced to gratify a personal enmity
must be the god which the applicant generally worships; and it is
imagined to be an extension of the protection granted by the god to his
worshipper’; which notion does not differ fundamentally from the
ancient Hebrew beseeching his god to destroy his enemies.

Witches are not thought of as possessing a power which is exclusively
and necessarily evil in itself; but all magical power is dreaded, because
it is susceptible of being used for harm. When such power is wielded by
one sex alone, this must inevitably arouse alarm in the other. Little
wonder, then, that men have sought to restrain women from using
magic, and have sought to acquire the secrets of die art in self-protec-
don. When magic comes to be exercised by organized male priesthoods,
the illicit practice of the art by women is regarded as presumably
malignant in intention. The magic woman who is no longer a priestess
must necessarily be a witch.

The Christian tradition that women brought death and sin into the
world is not peculiar to Christianity, but is universal. Father Lafitau
compares the First Woman of the North American Indians, who did
just this, to Eve, while Father Sahagun was struck by the same com-
parison in Mexico. Many African tribes regard the first woman as having
brought death into the world, as do the Eskimo and the Melanesians.
The Igorots of Luzon say that the first woman instigated men to fight.

We can obtain further clues concerning the significance of woman’s
mysteries by considering myths concerning the heavenly body with
whom they have always and everywhere been especially associated —
the moon, and to this topic we now turn.

CHAPTER 19

The Lord of the Women

Dangerous Character of the Moon

The dangerous character ascribed to women is extended, in the mind
of simple peoples, to that celestial body which is everywhere associated
with women — the moon. The tabu on menstruating women attaches to
the cause of menstruation also. In several of the myths which describe
how the first woman introduced death and woe into the world, she and
the moon are the same person — for instance, in the traditions of the
North American Indians. The Iroquois warriors, before an expedition,
consecrated themselves to the moon as to the spirit of relentless ven-
geance — a practice also observed by ancient Greek warriors. A Wyandot
tradition relates how Aataentsic, the first woman, planted fever-breeding
plants in order to destroy men. The myth calls up a picture of the witch
gathering magic herbs by moonlight.

In such a night

Medea gathered the enchanted herbs
That did renew old Aeson.

The natives of New Granada tell how a woman of surpassing beauty,
called Chia (i.e. the moon), appeared and taught doctrines contrary to
those received from their tribal god, and eventually caused a flood which
almost destroyed the human race. This myth has parallels in most parts
of the world: sometimes we are specifically told that the woman is the
moon, at others it can be inferred. Thus, in ancient Egypt mankind was
supposed to have been almost destroyed by fire and flood produced by
the moon-goddess Hathor or Isis. The prominence of the flood in most
mythologies is probably due not so much to dread of the damage which
floods cause as to their association with the moon, which is universally
held to control all waters.

In general, the moon is regarded as the mother of mankind. It has
even been suspected that Eve herself, whose name ‘Chawwa* means the
‘round one’, was originally the moon; a Rabbinical tradition represents
her as having at first consisted, like other lunar deities, only of a face. A
missionary once explained to a Syrian woman’s children that Adam and

THE LORD OF THE WOMEN 289

Eve were our first parents, but she protested against such new-fangled
notions, saying that our first parents were the sun and the moon.

The dangerousness of the moon is universally conceded. The Solomon
Islanders believe that she is always on the lookout to kill men, and the
Eskimo regard her as the cause of all plagues. The Tartars of Central
Asia regard the moon as inhabited by a giant who used to eat men, and
the peasants of modern Greece still regard the moon as anthropopha-
gous. Among the Maori the moon was called ‘the man-eater’, and was the
source of death; similarly elsewhere. The Bushmen throw sand in the
air and shout loudly when they see the new moon — their usual procedure
when they wish to drive away evil spirits. Other African tribes have
similar customs. One of the most familiar lunar superstitions is that it is
unlucky to see the new moon through glass. Obviously, in its primitive
form this myth could not have referred to so recent a luxury as glass
windows ; the myth was that it was dangerous for the new moon to enter
a house, or what was the same thing, to be seen from within the house.
Hence all savages come out of their dwellings to see the new moon. The
Bushmen are careful to build their huts in such a way that the moon
cannot shine in at the door, and in Louisiana the window shutters were
bolted at the new moon.

As we have seen, many uncultured peoples believe that in certain
phases of the moon it is inauspicious to undertake any work or enter-
prise; the new moon or the full moon being regarded as the most
dangerous phase. The malignant character popularly ascribed to the
moon reflects the predominance of lunar power in primitive religious
ideas.

Precedence of the Moon in Primitive Cosmology

‘Moon-worship naturally ranking below sun-worship in importance’,
wrote Tylor, ‘ranges through the same districts of culture.’ Tylor was
surprised to find that there were tribes who regarded the moon as a
deity, and ignored the sun, for he assumed that sun-worship took
precedence of moon- worship ; and that they normally co-existed. This
idea derives from the fact that people with whom we are most familiar
have reached the agricultural stage of development. Actually, moon-
worship long preceded any form of sun-worship and ranges through
different cultural areas.

Many students have reported this. ‘It is not the sun that first attracted
the savage’, remarks one writer; ‘we find moon-worship among most
utterly savage tribes in Africa and America; and it is noteworthy that
with them the moon is always regarded as a male, the sun as a woman;
not until later are those relations inverted. From this we may infer that
lunar worship is older than solar worship.’ Similar comments are made

T.M.— IO

THE MOTHERS

290

by Professor Hutton Webster, Dr Schultze, Dr Welckcr and many
others. Usener says: ‘Originally the moon was the only deity that was
worshipped.’ ‘As far as can be known, the veneration of the moon has
everywhere preceded the veneration of the sun’, observes Reclus. We
shall find similar conclusions with reference to particular regions and
peoples later. It is significant that innumerable popular superstitions
(which are relics of primitive conceptions) refer to the moon, while
scarcely any refer to the sun. The moon-cult survives to this day in our
countryside in the form of bowing and courtesies made to it.

The greater impressiveness of the sun (as it appears to us) is easily
explained away, if need be. The Nagas of Upper Burma, for instance,
believe that the sun shines by day because, being a woman, it is afraid
to venture out at night, whereas the bolder male moon is alone powerful
enough to face the darkness. It is pointed out that the moon is obviously
more powerful than the sun, for he commands the host of stars which
are his children, whereas the sun, if a monarch at all, is without sub-
jects. According to the Guaranis the light of the stars is derived from
the moon. Again, although the sun is more brilliant, we arc assured in
almost every primitive cosmological myth that this has not always been
so. The Methcis say that there were once two suns but they quarrelled,
and the wounded one became pale; while the Huitotos of Columbia hold
that the moon was once the sun and vice versa. These traditions persist
even in advanced religions. Thus in Brahmanical literature it is stated
that the sun ‘took to himself the moon’s shine’. Though in the Old
Testament the moon is ‘the lesser light’, in Talmudic literature, where
many ancient ideas are preserved, it is stated that originally the sun and
moon were equally brilliant. The Arabs also believe this, holding that
the angel Gabriel rubbed his wings against the moon and thus deprived
it of part of its brightness.

The suppositions which have been put forward to account for this
precedence of the moon are extremely inadequate. Thus Professor
Arrhenius suggests that, the difference between the seasons not being
so marked in the tropics as in temperate climes, the effects produced by
the sun have not been so noticeable to populations near the equator.
But in America, for instance, it is precisely among the inhabitants of the
arctic and antarctic areas that lunar worship is more pronounced, while
it is replaced by sun-worship in the equatorial region. Tylor, because he
was misled by the current theory that the religious ideas of primitive
man were moulded by impressions of natural phenomena, thought that
the sun must ‘naturally’ be regarded as more powerful.

In point of fact, primitive man is no more impressed with the spec-
tacles of nature than the boor who gapes at the enthusiasm of the
traveller for picturesque scenery, ‘Children and savages’, observes

THE LORD OF THE WOMEN 291

HofFding, ‘have as a rule no sense for the beauties of nature.’ ‘The
negroes of Congoland, as elsewhere in Africa, take surprisingly little
notice of the heavenly bodies or the phenomena of the sky.’

If the sun and moon were deified, it was because they were credited
with more or less power of influencing the welfare of men. The moon is
more important to the savage than the sun, because it is the marker of,
and therefore the cause of, time and change, and, in particular, of the
changes in women’s reproductive fife. And it is everywhere dreaded
because of its association with the sexual functions of women, and it is
regarded as the source of the awful powers ascribed to the witch.

The Moon as the Cause of Conception

As we have already noted, the moon is the regulator and cause of
menstruation, which is frequently regarded as being the result of actual
intercourse between the moon and women. Thus, for instance, among
the Murray Islanders, ‘the moon was supposed to be a young man who
at certain periods defiled women and girls, causing a bloody discharge’.
Menstruation is, according to many uncultured peoples, a form of
pregnancy; the foetus is supposed to be formed from the menstrual
blood. Pregnancy, as well as menstruation, is primitively considered to
be dependent upon or due to the moon. The Papuan natives, who ascribe
menstruation to the moon-god’s embraces, go on to explain that such
attentions aroused the jealousies of the husbands. He then appointed
that ‘in revenge all girls and young women should bleed when he
appeared, but the older and pregnant should be excepted, since in the
latter case he was responsible for their condition’. The Maori expressly
affirm that ‘the moon is the real husband of all women’, and that their
mortal husband is only, as it were, a subsidiary. The Gilbert Islanders
held that the first man and woman were forbidden to have sexual rela-
tions, the woman being exclusively reserved for the moon, who begat
children by her. The Hindus held a similar view. The Australian abori-
gines represent the moon as claiming that all women belong to him by
right. In Greenland, the Eskimo believe that the moon has intercourse
with their women, and ‘young maids are afraid to stare at the moon
imagining they may get a child by the bargain’.

The belief that the moon is essential to pregnancy is found among a
wide range of peoples, from the Indian tribes of Texas, where directly
after their marriage the women stand naked over a bucket of water which
has been exposed to the rays of the moon, to the Saorias of the Rajmaha
Hills, who believe that when the moon is absent from the sky, copula-
tion cannot result in pregnancy. ‘Among all negro races the moon and
generation are closely connected.’

In Central Europe it is believed that if a girl were to drink from water

THE MOTHERS

292

in which the moon is reflected and thus ‘swallow the moon’, she would
certainly become pregnant. In Brittany the women are careful not to
expose the lower part of their bodies to the rays of the moon, especially
in the first and last quarters when the moon is homed, for should they
do so they would at once conceive, or as they say, be ‘mooned’. In
Germany a pregnant woman must not linger in the moonlight lest she
bear a lunatic, as also in Iceland. In Africa the waxing moon is supposed
to produce male children, the waning moon female. In Cornwall there
is a saying, ‘No moon, no man’. In the Highlands of Scotland, girls
were wont to refuse to be married except at the full moon; and in the
Orkneys brides invariably visited megalithic circles, known as ‘the
temple of the moon’, and there prayed to the lunar power.

Moles, which are commonly known as ‘moon-calves’, are stated by
Pliny to be produced by women without sexual intercourse. The
Romans held a similar notion of the regulation of pregnancy by the
moon, and thought that satisfactory delivery depended upon its phases.
In ancient Egypt the fertility of women was thought to depend upon the
moon, and an inscription at Thebes states that ‘through his agency
women conceived’. The sacred bull Apis was held to be the outcome of
the impregnation of a cow by the moon. In Babylon pregnancy was
controlled by the moon, and the sex of the child was determined by
whether the moon had a halo or not at the time of conception. The
belief that a child born during the m oon’s eclipse w ould be born incom-
plete was held in Mexico and by Hind usanalWal ays .

The notion that the moon reveals to young girls their future husbands
is general in Europe. It is found in England, Ireland, Brittany, Ger-
many, France, Belgium, Portugal, Greece and among the southern Slavs.
In France, girls sing to the moon: ‘Lune, lune, belle lune, faites me voir
en mon dormant le mari que j’aurai de mon vivant.’ Girls among the
Indians of the Paraguayan Chaco address an identical request to the
moon. Among the southern Slavs the bridegroom at a wedding is some-
times called ‘Mr Moonshine’.

The Vedic view that the moon is the real husband of all women passed
over into Buddhism, and in a Buddhist legend Buddha himself was
begotten by the moon. In Persia the title of the moon was ‘the keeper of
the seed of the bull’, for, according to a very ancient myth, the primeval
bull or male principle of generation deposited his seed in the moon. In
ancient Babylon the moon-god, Sinn, is also a bull. The Greeks and
Romans retained the memory of similar views.

It is a world-wide custom for mothers to hold their new-born children
up to the moon, which is beneficial to the child. Sir James Frazer
regards this as an act of sympathetic magic, the growth of the child being
assimilated to that of the waxing moon. But actually among the Kashubs

THE LORD OF THE WOMEN 293

the waxing of the moon is regarded as causing weakness in the child ; it
is the waning which imparts strength. The true explanation is indicated
by the Isubu of the Cameroons ; when the mother shows the child to the
moon she says, ‘This is your grandfather’. And among the Kaffirs of
South Africa, when a mother presents a child to the moon, she says,
‘See, your child is growing’. So also among the Pelew Islanders.

It is true that conception is also ascribed to many other causes, e.g. to
eating the totem animal. But the totem or tribal spirit is frequently con-
founded in primitive thought with the cosmic deity or moon-god, and
usually identified with him. In Western Australia impregnation is
ascribed both to the totem animal and to the moon. Numerous other
agencies are credited with producing pregnancy, but many of these are
also emanations of the moon. Supernatural impregnation is often by
water — but everywhere waters are regarded as being under the control
of and derived from the moon ; or it takes place through the medium of
flowers, especially the lotus and its analogue, the lily. But these are
moon-emblems. In India the moon is called ‘the Lord of the Lotus’,
and in Egypt and Babylon the lotus was the emblem both of water and
of the moon deity.

Again, the sun frequently appears as the cause of impregnation — a
world-wide notion among agricultural peoples, but at this phase of
cultural evolution all the attributes, and especially the fertilizing attri-
butes, of the moon have, as we shall see, been transferred to the sun.

The Moon Primitively a Male

In primitive thought the moon is generally regarded a male. In primitive
mythology all spirits or gods are male or female as suits the occasion,
and primitive thought finds no incongruity in this. This does not mean
that the personified power is thought of as bisexual or asexual; on the
contrary. Primitive mythical conceptions are not a system of theology;
everyone is at liberty to regard a power as male or female, as circum-
stances demand. This is particularly the case with lunar deities, who
commonly have a male and female form. The female lunar deities do not
become prominent until relatively advanced phases of culture, and, in
particular, until the development of agriculture. In every instance it
would seem that the conception of the moon as male is the more ancient.
Thus among the Australian aborigines the moon is exclusively regarded
as a male; the sun, on the other hand, is female. The moon is pre-
eminently male in New Guinea and in Melanesia. In Polynesia the moon
is currently feminine, but in old myths it is represented as male. Among
the agricultural populations of Indonesia it is feminine but among the
more primitive populations it is masculine. It is male among the Eskimo
and the races of North-west America, and often remains male even after

THE MOTHERS

294

the development of considerable agricultural activities. It is still male in
Mexico and among the Caribbean races of Central America, and among
most, if not all, of the tribes of South America. It is male among
Mongolian tribes and Tartar populations, and among the Japanese. With
the Chinese the moon is a goddess, but in popular tradition it is repre-
sented as a man. It is m§lgjnj[nx^^ and among the Nagas,

the Todas, the Khasis, the Shans, the Siamese; as also among the
ancient Persians and the Armenians. In all Semitic languages i the moon
is masculine and the sun feminine. The samels true among all the more
primitive tribes of Africa, among the Slavs, the Finns, in Scandinavia
and in Iceland, as well as among all Teutonic races. The moon was
male among our own Anglo-Saxon ancestors, and the word remained
masculine in English up to the time when arbitrary genders disappeared
from the language. It still remains male in English folklore, which knows
only the man in the moon. Our habit of regarding the moon as female is
due only to our training in classical mythology — but even in Greece and
Rome the moon was originally masculine. It was universally masculine
among the Celts. In France, where classical tradition has been prolonged,
the moon is feminine, but among the Bretons and the Basques the
original sex reappears.

The Moon as a Source of Magical Power

But while the moon, as ‘the real husband of all women’, is thought of as
a male, it is at the same time associated with the functions, not of men
but of women. It is the source, not only of their reproductive powers
but of all their other powers, and especially of their magical powers.

The moon is everywhere regarded as the source of magic; the ‘Lord
of the Women’ is also the Lord of witches and magicians. Thus among
the North American Indians the moon was ‘the chief of manitus’, or
wizards. Among the Eskimo the power of the ‘angakut’ derives from the
moon, and when he is in a trance his spirit is transferred to it. A Chukchi
shaman ‘when he desires to make especially powerful incantations, must
strip himself naked, and go out of his house when the moon is shining’
to invoke the moon. The Tartars of Central Asia consider that women
can practise witchcra ft only once a month, at the new moon; and in India
witchcraft musTBeperformed by moonlight.

Many other instances can be given. T3ntEe Gold Coast the same word
means both moon and witchcraft. In ancient Greece all witchcraft was
held to derive from the moon, and the moon-goddess, Hekate, was the
special patroness of witches. It is still held by Greek peasants that no
witch can work without the aid of the moon. In the Shetlands a witch,
to strengthen her powers, lay for hours in the moonlight to become
thoroughly saturated with its influence. Among the survivals of pagan-

THE LORD OF THE WOMEN 295

ism especially condemned in the Carolingian capitularies is the belief
that women can exercise witchcraft by means of the moon. In Germany
deeds of witchcraft are particularly looked for on Mondays. Cornelius
Agrippa, to reconcile this view with astrological doctrine, said that,
though magical power came from all heavenly bodies, it could only be
transmitted to the earth and its inhabitants by the intermediary of the
moon.

The fascination which woman exercises, her beauty (which many
uncultured peoples regard as a form of witchcraft) is also bestowed by
the moon. The Vedas state that woman’s beauty is derived from the
moon, and is promoted by eating the flesh of the moon-hare, or by
applying preparations obtained from it. The beauty of women is ex-
pressed in all Oriental languages, as well as in Southern Europe and in
Polynesia, by assimilating it to the moon. The moon-goddesses are the
goddesses of beauty and love.

The faculty of prophecy is likewise derived from the moon. ‘The
ancients’, says Lydus, ‘regarded the moon as the leader in all divina-
tion.’ In primitive Greece and among the Semites all oracles originally
derived their inspiration from lunar deities.

The Moon as Cause of Time and as Destiny

The association of the moon with prophecy derives from the fact that
throughout primitive culture the moon is the only measure of time. The
solar year is a relatively late discovery, made possible only by ingenious
astronomical observations. The cycle of seasonal changes is, of course,
visible to all peoples living in temperate climes, but this affords no fixed
point which can serve to measure duration. Throughout the greater part
of human development the moon, whose name in our language is from
the root ‘mas’, measure, ^ensura’, has been the sole marker of time.
Even in cultures so advanced aT Islam the sun is never thought of as
affording a measure of time. That function belongs to the moon. ‘God
created the moon and appointed its ‘houses’, 1 ” says the Koran, ‘in order
that men might know the number of the years and the measure of time.’

But in the primitive mind the function of the moon is not simply to
measure time but to create it. The word moon is not derived from
‘measure’ — the word ‘measure’ is derived from the name of the moon.
It is the cau se ofj time, just as it is the cau se of menstruati on. Hence the
moon stands for the conception of fate or destiny which pervades the
thought of uncultured humanity.

Further, the moon, as we shall see in greater detail, stands in primitive
thought for perpetual ren ewal , immortality, eternity. The Siouan tribes
of America, for example, called her ‘tKe Old Woman who never dies’.
Her name, Aataentsic, among the Iroquois tribes means ‘the Eternal

THE MOTHERS

296

One’. In Polynesia, the moon was regarded as possessing the secret of
immortality; she was perpetually renewed. Among the Chams, the
moon has the gift of eternal youth; she is never more than thirty.
Similarly among the Chinese. Among the ancient Egyptians the moon
was ‘the maker of eternity and creator of everlastingness’. In Latin
inscriptions her epithet is ‘the eternal’. In Russia she is ‘the deathless
one’.

In primitive thought the eternal time-creating nature of the moon
imparts to it an inexorable character, setting it above all other powers.
The Iroquois recognized that there was a power above that of all
‘manitus’, whose decrees were unalterable. This conception, which has
its parallel in all primitive mythologies, corresponds to that of the Greek
Moira, who stood above the gods. But the Moira, or the three Moirai,
or Fates, were originally the moon. ‘The Moirai’, says Porphyry, ‘are
referred to the power of the moon’, and in Orphic writing they are
spoken of as parts of the moon. Destiny is assimilated by Sophocles to
the moon. The Arabian goddess, Manat, was the moon and was also
Fate. The Teutonic Norns, like the Greek Moirai, were moon-goddesses
and were ‘older than the gods’.

Mr Comford has shown the attributes of the Greek Moirai to be the
prototype of the scientific conception of natural law; he points out that
the Greek conception of Moira, or Fate, is essentially that of a dividing
divinity and measurer of lots, who apportioned his special sphere of
activity and functions, not only to every human being but also to every
god. These are likewise the functions of the primitive mother, who
allots to each his portion of food, and who, in early agricultural phases
of culture, is also the apportioner of cultivable land.

Lunar Divinities commonly Triune

Like the Moirai, lunar deities are usually threefold. Among many
uncultured peoples the waxing, the full and waning moons had three
different names and were conceived as three different persons. Thus in
New Britain the moon consists of the ‘White Woman’ and her two sons,
the waxing and the waning moon. Among the natives of Northern
Ashanti ‘it is well-known that the satellite is inhabited by three beings,
similar to men in appearance but provided with enormous ears which
completely cover their faces. One of these is white and the other two
black’. In Dahomey the badge of moon-priestesses consists of a white
shell and two black beads. The Indians of New Granada represent their
lunar deity by three crosses, one large one flanked by two smaller ones.
The month was divided into three parts among the Germans and the
Celts, as likewise among the Greeks, the Romans and the Semites.

Countless triads of Greek goddesses, such as the three Charities, the

THE LORD OF THE WOMEN 297

three Horai, the three Syrens, the three Hesperides, the three Gorgons,
the three Erinyes, are primitively scarcely distinguishable from one
another or from the Moirai. The Muses were originally also three in
number, and were deities of the night heavens, governing the stars.
The Mothers, Nurses or Nymphs, who bring up infant gods, are also
forms of the Moirai, who preside over the destiny of every child, and
were, like them, threefold. These triads of Hellenic goddesses were
regarded at will as one or three. They were triune, or three in one. Like
them, the great goddess of the Semites was worshipped at Mecca in
threefold form as three sacred trees, and was spoken of as the Three
Virgins. In Phoenicia and Carthage, as in Krete and ancient Greece, the
great goddess was represented by three pillars. The Jewish god, Yah-
weh, appeared in the form of three men whom Abraham addressed as
one. Threefold deities are prominent among the races of Northern
Europe and among the Celts. Thus Brigit, the Norns, the Walkyries
had the threefold character. They were impersonated by three priestesses
who officiated over the birth of every child. They became the three
weird sisters. The three Fatal Sisters survive in popular tradition as the
three fairies, or fays. In many of the stories in which they figure two of
them are deformed and one of them is rounded and beautiful like the
full moon. Just as the moon-god of the Ashanti is black in two persons
and white in the third, so the Erinyes were similarly two parts black and
one white.

The principle by which the threefold division of the month was applied
to lunar deities has sometimes been extended to solar and other deities.
Thus Zeus has been described as having three hundred heads, corres-
ponding to the three hundred days of the year. More frequently, how-
ever, the threefold character of lunar deities having once set the pattern,
other deities are represented as triune. Thus Zeus was represented at
Corinth with three eyes. Similarly the ancient Mexicans worshipped
their gods as a trinity denoted by three crosses. The heathen Slavs
similarly represented their deity with three heads. The Nordic gods were
worshipped at Upsala as a trinity.

The profane mimicry of the mysteries of the Christian religion by
heathens has naturally been a cause of annoyance to missionaries. Tt is
strange’, says Father d’ Acosta, ‘that the Divell, after his manner, hath
brought a trinitie into idolatry, for the three images of the Sunne called
Apomti, Churunti, and Intiquaoqui, which signifieth father and lord
Sunne, the sonne Sunne, and the brother Sunne. ... I remember that
being in Chuquisaca, an honourable priest showed me an information,
which I had long in my handes, where it was prooved that there was a
certaine “huaca”, or oratory, whereas the Indians did worship an idoll
called Tangatanga, which they saide was one in three, and three in one.

T.M. — 10*

THE MOTHERS

298

And as the priest stood amazed thereat, I saide that the Diveil by his
infernall and obstinate pride (whereby he alwayes pretendes to make
himself God) did steale all that he could from the trueth to imploy it in
his lyings and deceits. 5

Pagan symbolizm has been extensively adopted in Christian myth.
Thus the three Moirai, or Mothers, have survived as the three Maries,
the three daughters of Holy Sophia, who is stated to be the moon, or
as Faith, Hope and Charity. In a shrine at Vallepietra, near Anagni, a
threefold Christ is held in high repute (Fig. 1). Even cruder ikons were
common until lately, representing God with three heads, or with three
faces (Fig. 2). The Church did its best to put down this disguised

heathenism; Pope Urban VIII had a number of three-headed gods
removed from the churches of Italy, and in 1628, caused these Holy
Trinities to be publicly burned.

The Moon as the Source of Lunacy

Lunacy, ‘demoniac phrenzy, moping melancholy and moon-struck
madness 5 , are universally regarded as caused by the moon. Mental
derangement is also a qualification for the prophetic, magic or saintly
character. The powers of the prophet, magician or priest are thought to
depend upon inspiration or possession, and are manifested by various
nervous phenomena — convulsions, trances, hypnosis, somnambulism.
The word ‘shaman 5 , like the Greek word ‘mainad 5 , means ‘the raging
one 5 . Aspirants to prophetic and magic powers seek to cultivate nervous

THE LORD OF THE WOMEN

299

hyperaesthesia and instability. In Egypt, ‘they look upon all madmen,
imbeciles, lunatics, and such as are afflicted with the falling sickness, as
saints’. Lunacy is regarded with reverence throughout the East. The
ancient Hebrews held that ‘the prophet is a fool, the man that hath the
spirit is mad’. The Baralonga honour demented persons, ‘believing
them to be under the direct influence of their tutelary deities’. ‘Regard
for lunatics is a universal trait among the American tribes.’ ‘The associa-
tion of poetry, prophecy and idiocy’, observes Sir John Rhys, ‘is so
thoroughly Celtic as to need no remark.’

Not insanity alone, but every disease involving convulsions, is ascribed
to the moon. Epilepsy, ‘the sacred disease’, was formerly spoken of as
lunacy. Jesus, for instance, cured ‘lunatics’, oefoiviaCo/tevoi, which
word is translated in the R.V. as ‘epileptics’.

Lunar Animals: the Dog , the Hare , the Cat

Hekate’s attendant animal was the dog, which worshipped her by
barking at the moon, and announced her coming by howling. The
Peruvians, during eclipses of the moon, used to beat their dogs to make
them howl, ‘thinking her affectioned to doggcs’; so did the Hurons and
Iroquois, while the Apache honoured the moon by yelping. Sometimes
an Indian, having shot a dog, found the corpse of a woman.

The association in primitive thought between witches and the moon
is illustrated by the connection of both with certain animals. It is a
widespread notion that witches may assume the form of a hare. ‘The
hare’, says Mr Henderson, ‘is the commonest disguise of the witch in all
the northern countries of Europe.’ He has ‘personally heard of and
known many women who were regarded as having the power of shifting
themselves into hare-shape’. In Wales, ‘only the women can become
hares’. In the Isle of Man witches regularly become hares ; there is one
instance of a man doing so, but he was a smith and thus a warlock. In
Scotland a witch-hare cannot be hit with ordinary shot, but can be
wounded with a crooked sixpence. In many stories found all over the
world the wounded animal enters a cottage, and the hunter, following,
finds there a bleeding, wounded woman.

The transformation is sometimes effected in a manner reminiscent of
totemic rituals ; namely, by the witch anointing herself with hare’s fat.
Indeed, hare’s fat is believed to bring out the witch-nature of any
woman, as was believed in medieval Europe. Dalla Porta describes how
modest women ‘set a lamp with characters graved upon it, and filled it
with hare’s fat; then they mumble forth some words and light it. When
it burns in the middle of women’s company, it constrains them all to
cast off their clothes, and voluntarily to show themselves naked unto
men; they behold all their privities which otherwise would be covered,

THE MOTHERS

300

and the women will never leave dancing so long as the lamp burns’.

For a hare to cross one’s path is regarded as most unlucky. In Scot-
land, Ireland and Brittany the very word must never be mentioned.
Sailors believe that the presence of a dead hare on board ship will cause
bad weather. In France to eat hare’s flesh is regarded as making women
more attractive; that is, imparting to them powers of bewitching men.
Hare’s flesh was eaten for this purpose in ancient Rome. Eating hare’s
flesh is also believed to render women fertile. According to a widespread
belief among the Jews, Greeks and Chinese, all hares are female and are
impregnated by the moon. The antiquity of their association with witch-
craft among the Celts is testified by the fact that Queen Boadicea, when
opposing the Roman armies, drew a hare from her bosom and followed
its guidance in directing her attacks.

The hare is also widely associated with the moon throughout India,
where the moon is called the ‘bearer of the hare’. In Tibet and among
the Mongols of Central Asia, as in China and Japan, the spots in the
moon are interpreted as a hare, put there by a great magician, a myth
which is found as far south as Ceylon. The hare plays a prominent part
in folklore throughout South Africa. ‘He is a small creature, but with
one exception all the animals are as clay in his hands’; and is often
regarded as the messenger of the moon, who pronounced upon mankind
the doom of mortality.

Such myths are equally popular among the Indians of North America,
but the hare is often substituted by the rabbit, never clearly distin-
guished from the hare. There can be little doubt that the Great Hare,
the most familiar form of the Great Manitu among the Iroquois
nations, was originally the moon. Hekate was a hare, and in Karia was
worshipped as the hare-goddess, and her chief shrine was at Lagina, the
hare-city. Artemis was a hare and hares were sacred to her. Even in the
present day, in some parts of Albania they ‘consider it a sin to kill a
hare’. Dionysos, who was a moon-god, changed himself into a hare.

The close association of the hare with witches and with the moon is
found in Northern Europe equally. The part which the hare played in
the agricultural cult of the ancient Germans is indicated by the impor-
tance attached to it in connection with Easter. It is the Easter-hare
which lays the Easter eggs, and it figures in association with them in
Easter cards. Much circumstantial evidence indicates that it occupied
an equally important place in the Celtic cults of Great Britain and Ire-
land. The custom of hunting the Easter-hare was kept up until lately
with much pomp by the mayor and aldermen of Leicester. We know that
the hare was tabu as food to the ancient Britons, and that it was eaten as
a sacred food by the kings of Ireland. The hare upon which Boadicea
relied was probably the sacred animal of her goddess.

THE LORD OF THE WOMEN 301

Why is the hare so universally associated with the moon ? There can
be little doubt that its original relation to the lunar deity was sacrificial.
In the cults which belong especially to women, the animals sacrificed
are almost invariably small animals, such as birds, dogs, rabbits, hares
and pigs, for women are seldom able to obtain the large animals used in
sacrifice by hunters and herdsmen. Furthermore, the moon’s fructifying
powers are paralleled by the fertility of the rabbit or hare, which in Aztec
hieroglyphics had the same ideogram as does the earth.

In West Africa the cat is the commonest animal into which a witch
may change, and was the usual disguise of the familiar spirit of witches
in Western Europe. The cat is accordingly credited with having nine
lives, with being able to see in the dark and with being able to cause
storms. It is a widespread superstition in Europe that a cat washing
herself behind the ears forebodes rain, while in Asia Minor rain is
expected if a cat licks her paws. In Indonesia to compel a cat to wash
herself is considered a good way to bring about a good downpour ; while
the natives of southern Celebes, when there is a drought, carry a cat
round the fields in a sedan-chair, squirting water over it. In the Middle
Ages heretics were commonly accused of worshipping cats. The Cathari
were said, erroneously, to have derived their name from the practice,
and the charge was brought against the Knights Templar. The Stad-
inghi, a heretical sect of the thirteenth century, were accused in a Papal
Bull of the crime of keeping black cats.

The important part played by cats in the ritual of witches ‘is clearly
derived from an early form of sacrifice’. Shakespeare appears to refer to
the practice, and the sacrifice of cats to produce rain is widespread. In
Bohemia and in Russia a black cat is buried alive in a field to promote
the fertility of the land ; and in some parts of Lancashire it is still thought
that to shut the cat up in the oven promotes good luck. Cats are sup-
posed to become fat or lean with the waxing or waning of the moon, and
Cornelius Agrippa says their eyes grow wider or narrower according to
its phases. They are ‘lunar animals, and are of the same nature as
menstrual blood, with which many wonderful, miraculous things are
wrought by magicians’. The Australian aborigines and Tasmanians
identify the moon with a wild cat, and the North American Indians
believe that there is a cat in the moon. The cat was sacred to the goddess
Freija, who rode on a chariot drawn by cats; and in Egypt cats were
sacrosanct. So abundant are cat-mummies in Egypt that shiploads of
them have been brought to Europe to use as manure. According to
Plutarch, the cat in Egypt denoted the moon. ‘Though this may look
like a fiction, yet there can be no doubt that the eyes of cats seem to
grow larger at the full moon, and to decrease again and diminish in
brightness upon the waning.’

3°2

THE MOTHERS

The Moon as Spinstress

All feminine attributes, not only fertility and magic, are dependent upon
the moon. One of the activities most commonly ascribed to it is spinning
and weaving. The moon is a spinster in innumerable folk-tales in
Germany and Italy, and the same idea is found among the North
American Indians, in Central Brazil, in the Banks Islands and elsewhere.
In Borneo the moon assumes the form of a spider. In ancient Egypt, the
moon-goddess, Neith, invented weaving. Artemis, Athene, Aphrodite
and the Nymphs were spinstresses, as was the Nordic goddess, Freija,
and the Teutonic goddesses, Holda and Bertha. The hare is also found
in folk-tale as a weaver. Jewish women did their spinning and weaving
by moonlight, while German women held that spinning must on no
account be done by moonlight. The Dayaks believed a woman could not
weave beautiful patterns without the inspiration of her divine protectress.

It is natural enough to connect the moon as a measurer of duration
with the spinning of the thread, or the weaving of the web of time and
destiny. Nevertheless, the primary connection between such pursuits
and the moon was probably not poetical, but more direct, for the moon
is engaged in almost every other occupation belonging to primitive
women. It is known as a maker of bark-cloth, as a basket-maker, pot-
maker, grinder of seeds, maker of fires and as a cook. Among the Eskimo
the fires of the sun are stoked up by the lunar deity.

In Europe the man in the moon is most commonly shown as carrying
a load of firewood, which is interpreted as a punishment for Sabbath-
breaking, and in some popular myths he is Cain or Judas Iscariot. The
representation, however, is much older than Christianity. The Yoruba
of the Gold Coast see a Sabbath-breaker in the moon who carries a load
of firewood, and the same idea is found among all the southern Bantu.
The firewood of the man in the moon has baffled many. But the gather-
ing of firewood is simply one of the occupations of primitive women.
Among the Banyoro the first act of a bride, after marriage, is to draw a
small pitcher of water from a well and to gather a small bundle of sticks.
‘The drawing of water and the carrying of fuel typified’, says the Rev.
J. Roscoe, ‘the duties of the wife.’

The Moon regarded as the Producer of Vegetation

In tradition, there are trees and shrubs in the moon. In some repre-
sentations the moon-man is surrounded by thorn bushes. Thom bushes
were sacred among the Celts and Teutons, and were used for cremating
the dead. Trees are observed in the moon in various Molucca Islands,
in Melanesia, in the Malay Peninsula, in China and in Sweden. In
modem Greece the man in the moon, who is Cain, is engaged in cutting

THE LORD OF THE WOMEN 303

down a tree which grows again every month. The same notion is current
in China.

One of the most prominent attributes of the moon is, in fact, that of
producer of all vegetable life : it is known among the Brazilians as the
‘Mother of Vegetables’; among the agricultural tribes of North America
as the ‘Mother’ of corn, and is assimilated by the Cherokee to the maize
itself. This notion is also found in many other parts of the world. Thus
in India, the moon is known as ‘the bearer of seed, the bearer of plants’,
and is identified with some, ‘the Lord of Plants’. The Caribs picture
the world as having arisen from a gigantic tree, resembling the world-
tree of Nordic mythology, Yggdrasil, which in the Voluspa is called ‘the
tree that metes out the fate of men’. It is also the Tree of Mimir, who
appears to have been an ancient Nordic moon-god. All Semitic moon
deities were similarly associated with trees and sacred bushes, while in
the Admiralty Islands the moon is identified with the coconut-palm. On
some Babylonian cylinders, the moon-god is represented as the trunk of
a tree with branches growing from every part of his person. In Krete and
archaic Greece also lunar deities were worshipped as trees.

The moon is particularly identified with the juices of vegetables and
the sap of trees, which is supposed to rise and fall with the phases of the
moon; hence the rule that timber must be felled at the waning of the
moon. The juices of vegetables arc regarded as their life-blood or soul,
and so as the life-blood or soul of the moon. Accordingly the highest
forms of divine inspiration are attained by drinking the juice of the
soma, or, in South America, the concoction known as chicha, or the
juice of the grape, which was regarded in Western Asia as the blood of
the deity. To chew the leaves or fibres of the lunar plant is a necessary
preliminary to divine and prophetic inspiration. Among the Semites,
those trees which exude thick saps of aromatic gum were regarded with
special reverence, and the fluid was employed as incense. ‘The value of
the gum acacia as an amulet’, says W. Robertson Smith, ‘is connected
with the idea that it is a clot of menstruous blood, i.e. that the tree is a
woman.’ The moon was thus the divine counterpart of the primitive
cultivator of the soil. Hence the presence of trees and shrubs in the
moon.

The Moon as Controller of Moisture

The moon has, by universal consent, control of all water and moisture,
as part of her office of bringing forth vegetation. Primitive women
secure the fertility of the earth by watering it and by performing rain-
making incantations, in which they are assisted by the moon. According
to the Eskimo, snow comes directly from the moon. In British Columbia
the moon is represented as bearing water-pitchers. The Indians looked

THE MOTHERS

3°4

upon the moon as being the actual cause of rains and gods, and among
the Algonkin tribes the terms for moon and water were the same.
Similarly the serpent goddess of Mexico, who was the moon, was also
the goddess of water and the ocean. The moon is represented in Maya
manuscripts by a pitcher of water. The idea that the moon is ruler over
waters is found in China and Japan, in Central Australia and Central
Asia, in Brazil and in many other parts of the world.

Frogs who croak to the moon never fail to obtain rain from him. In
the Ganges valley, the women, when there is a drought, slowly crush a
frog to death, and its croaking is believed to be an infallible rain-charm.
The frog, often interchangeable in myth with the toad and with the
dragon, is a universal emblem of water and the moon. Among the
Pueblos, the frog-clans are the rain-making clans, and the great goddess
of Mexico, who is the moon and the ruler of waters, was represented by
a huge emerald frog. The North American Indians see in the moon the
‘Primeval Toad’, which contained all the waters of the world and caused
the flood by discharging them. Many races, e.g. the Chinese, see a frog
in the moon. The toad is also regarded as an emblem of the womb; and
in Germany women suffering from uterine troubles present images of
toads to the Virgin Mary. In Semitic mythology the moon deity is
invariably the controller of water and the primal ocean out of which all
things arose. So, also among the Greeks and the Romans, the moon was
regarded as the source of all moisture. This list could be lengthened
almost indefinitely: medieval writers supported the idea. ‘Water’, says
Cornelius Agrippa, ‘is the lunar element, the water of the sea as well as
that of rivers, and all things humid, the humours of trees and of animals,
and more especially those humours that are white, such as the white of
eggs, fat sweat, pituitary discharge and superfluities of the body.’

The notion that the moon is ‘the governess of floods’ persists among
ourselves in the notion that the phases of the moon are related to changes
in the weather, an idea which persists in spite of repeated demonstra-
tions of its falsity. It is anything but obvious why on grounds of mere
nature symbolism the moon should be universally associated with
moisture. The connection between the moon and the tides appears to be
unknown to most uncultured peoples. Those daring seafarers, the
Polynesians, had no such notion and thought them due to the presence
of a huge monster which breathed twice a day. If any connection be-
tween the tides and the moon is recognized by some uncultured
peoples, probably it is deduced by them from the moon’s character as
ruler of the waters, and not the other way about.

On the other hand, nothing would be more logical than that the
source of magical power should be the controller of the waters upon
which fertility depends. Woman is the primitive water-carrier and also

THE LORD OF THE WOMEN 305

the primitive rain-maker. The two functions, practical and magical, are
scarcely distinguished from one another. The drawing of water and the
watering of the fields are regarded as magical operations, and are carried
out with the solemnity of a ritual.

In short, the attributes of the moon in primitive thought are not the
products of poetic symbolism, but are the transferred characters and
activities of primitive women, which are regarded as being derived from
and controlled by the magic power of the moon. Everywhere — from
China to Peru, from Ticrra del Fuego to the Arctic — the moon is
regarded as the counterpart of women, their special deity, the controller
of their beings and the source of their powers. ‘The nature of women*,
as Rabelais says, ‘is figured by the moon.*

Now, just as the functions of the moon-deity reflect woman’s share
in the sexual division of labour, so also the moon’s supremacy as the
primitive cosmic power is the counterpart of the magical powers of
primitive woman. Hence wherever we find the moon pre-eminent, we
are entitled to conclude that women are, or were, pre-eminent too. This
is perhaps why Spartian says: ‘All the learned, especially in Harran,
believe that those who honour the moon as a female deity and give it a
feminine name remain for ever enslaved to their women, but those who
worship it as a male deity and give it a corresponding name, rule over
their women.’ Likewise, according to Palgrave, our Anglo-Saxon
ancestors ‘had an odd notion that if they addressed that power as a
goddess, their wives would be their masters’.

CHAPTER 20

The Resurrection and the Life

The Serpent and Eternal Life

With the fall of man and the origin of death, with woman and the
moon, who are held responsible for these disasters, is associated the
serpent, an animal which plays a larger part in religious myth than any
other. In all symbolism it is the emblem of immortality because, instead
of dying, it changes its skin. Uncultured peoples hold the serpent to be
the only animal (together with lizards and other reptiles, and also crabs,
which are regarded as equivalent) which possesses the gift of immor-
tality. Savages, it should be remembered, do not regard death as a
natural event, but ascribe it to witchcraft — woman’s witchcraft, or,
what is the same thing, the moon’s. True, the effects of old age are
recognized in the wrinkling of the skin ; the natives of British Columbia
consider that men die because their skin is too soft. If men could change
their skins as serpents do, they would also be immortal.

The belief that the first human beings preserved their immortal
nature by this faculty, and that serpents were involved in its loss, is very
widespread. Thus the Tsimshian of British Columbia relate that their
tribal hero, or first man, applied to the moon-god for the gift of renewed
youth. The divine physician purified him by bathing him, and his skin
fell from him in the form of scales, leaving him as white as snow. The
tribes of central California say that death was introduced among men
by the machinations of the lizards, who obtained from men the power of
changing their skin; before this men did not die, but merely moulted —
a view also held by the Caribs. The ancient Mexicans also had a god,
Xipe Totec — originally a moon-god, but regarded later as a god of
vegetation — who regularly renewed his youth by putting on a new skin.
Again, the Arawak of the Orinoco tell how a boy was swallowed by a
serpent and re-emerged, to the admiration of his friends, with a new
skin, which was mottled like a serpent’s. We come upon an almost
identical story in Australia.

Some Australian tribes believe that the dead put on incorruptibility
by an opposite process ; instead of being swallowed, they themselves cat
a serpent. In New Caledonia the serpent is replaced by a sea-monster,
but fishes and serpents are often interchanged in popular conceptions.

THE RESURRECTION AND THE LIFE 307

In Tonga the god Tangaroa, who is a lizard, is said to have sent his
two sons with their wives to people the island. One of those sons was all
that a son should be, but the other turned out badly. So Tangaroa
appointed that the good son and his descendants should have white
skins and should go to the land of the immortals; but the bad son and
his people were condemned to have black skins and to remain in Tonga.
The same story is told in Samoa, but the personages are lizards.

In the languages of Melanesia the term for ‘to slough one’s skin’ is
equivalent to ‘living for ever’. In the New Hebrides the origin of death
is accounted for in the following story : an old woman went down to the
sea and, casting off her old wrinkled skin, as lobsters cast their shells,
she became a young and beautiful woman. But her grandchild failed to
recognize her in her disguise, and would not let her take her in her arms.
So she returned to the sea and put on the old skin. The same story is
told with unimportant variations throughout most parts of Melanesia;
we find it again in Papua.

The myth takes another form in the New Hebrides, in the island of
Ambryn. The two gods, Barkolkol and Buglian, who respectively repre-
sent the bright aspect and the dark aspect of the moon, created man.
When man began to age and his skin grew wrinkled; Buglian said to
Barkolkol: ‘Our man is getting wrinkled, what shall we do ?’ Barkolkol
replied : ‘We will skin him like an eel ; he will grow a new skin like the
serpent, and thus he will be made young.’ But Buglian said: ‘No. When
he is old and ugly, we will dig a hole in the earth, and put him in it.’ In a
similar story from the Gazelle Peninsula of New Britain the good god
Kambinana said to his wicked and foolish brother Korvouva: ‘Go down
to the men and tell them to cast their skins, so shall they avoid death.
But tell the serpents that they must henceforth die.’ But Korvouva, the
fool, reversed the message. The Annamites also regard mankind as
having been defrauded of immortality by the serpents, who so terrified
the god’s messenger that he reversed the god’s message.

The same notions are widely current throughout Africa. In the forests
of the upper basin of the Congo, an account of the origin of death is
current similar to that which is so widespread in Melanesia. The first
man, it is said, had two wives. In course of time their skins began to get
wrinkled. The elder wife then went to a hut, and proceeded to take off
her old skin, and laid it down on the winnowing fan. Unfortunately the
second wife happened to need that implement and went to the hut to
fetch it; there she saw her co-wife radiant with renewed beauty. But the
process of renewal had not been quite completed, and her entry broke
the spell, causing both women to fall dead ; death was thus introduced
into the world. The serpent, on the other hand, never dies, and cannot
be killed unless it is completely crushed.

THE MOTHERS

308

This myth of a disastrous error or fraud is common. Among the
Gallas, tradition asserts that the order to slough their skin was by mis-
take delivered to the serpents instead of to the human race. Similarly
the Wafipa and Wabende of the shores of Lake Tanganyika relate that
God once asked the creatures of the earth, ‘Who wishes not to die ?’
Unfortunately all were asleep except the serpent, which answered, T do.’
The Sea Dayaks have the same story.

In the more common African form of the tradition the lizard is the
chief means of defrauding mankind of the gift of immortality. Thus, for
example, among some Basuto tribes it is related how the good king
Leobu, grieved at the sufferings of his people, decided to send his own
beloved son to announce the good news that they should not die but
have life everlasting. The king’s instructions were, however, overheard
by a professional runner, whose name was Khatoane — that is, the Lizard
— and he ran to the people and gave them the opposite message. In a
Zulu version, the chameleon is sent with the message ‘Let not men die’,
and the lizard with the message ‘Let men die’. The lizard naturally
arrives first. This story, in slightly varying forms, is current throughout
the greater part of Africa, from the Cape to the Sahara.

Identical conceptions were familiar to the ancient Greeks. The term
by which they spoke of the cast skin of a serpent was yijQa c, that is to say
‘old age’. Aelian tells how Zeus, taking pity on man, decided to bestow
immortality on him, sending the gift on the back of the donkey. The
donkey, however, was defrauded of his burden by a serpent in exchange
for a drink. In another story, the wise Polyidos brings about the resur-
rection of Glaucus, the infant son of King Minos, who died from falling
into a cask of honey, by rubbing the child with certain herbs, with
which he had seen a snake resurrect a dead companion. In another
version, Glaucus obtains possession of the herb of immortality — from
the moon-hare. (The idea that the moon-hare possesses this herb is also
found in China.)

The ancient Egyptians likewise associated immortality with the
serpent. In a passage of the ‘Book of the Dead’, the deceased prays that
he may become like a serpent. ‘I am the serpent Sata, whose years are
many’, he says; ‘I die and I am bom again. I am the serpent Sata which
dwelleth in the uttermost parts of the earth. I die and I am bom again,
and I renew myself, and I grow young each time.’

In the Gilgamesh epic, one of the most ancient literary monuments
of the world, a story exactly similar to that of Glaucus is told. In
Genesis, which appears to have been largely inspired by Babylonian
myths, the serpent, as in the traditions of the whole uncultured world,
defrauds the human race of immortality. But the Biblical narrative has
been adapted to later conceptions of Jewish theology. A Rabbinical

THE RESURRECTION AND THE LIFE 309

commentary on Genesis iii. 14 explains that Adam, after the Fall, did
not possess the same skin as before, his first skin having presumably
been immortal; it is in consequence of thus having changed the skin of
the first man that the serpent is compelled to moult his. An early Mus-
lim doctor, drawing, no doubt, from the same traditional sources, gives
further particulars. ‘When the fruit had descended down Adam’s throat
and reached his stomach, the skin which Adam had in Paradise fell from
his body. That of Eve fell off likewise, and the soft flesh of their bodies
remained exposed as it is with us at the present day; for the skin which
Adam had in Paradise was similar to the substance of nails. When it
became detached, only a small portion of it remained on the tips of the
fingers ; and thereafter whenever Adam or Eve beheld the nails of their
fingers, they were reminded of Paradise and its delights.’

The Redeemer Himself was likened by the apostle to the serpent : ‘As
Moses lifted up the serpent in the wilderness, even so must the Son of
Man be lifted up : that whosoever believeth in him should not perish,
but have eternal life’ (John iii. 14-15).

Later, the notion that the serpent possesses the gift of immortality
was a definite tenet in Europe among the alchemists and practitioners of
occult arts. Serpents, being undying, are accordingly widely regarded as
the spirits or souls of the departed.

Immortality Dependent on the Moon

The serpent thus shares the moon’s gift of immortality. The moon itself
is thought by some peoples to owe its faculty of rejuvenation to the
power of casting off its old skin. The ‘old women’ and other personages
who, in the Melanesian and Papuan myths, renew their lives by slough-
ing their skin are, in fact, personifications of the moon. In the languages
of Melanesia the only way of expressing the notion of eternity is the
phrase ‘ul ta marama’; that is to say, ‘casting off one’s skin like the
moon’. So when the missionaries desire to explain that God is eternal, they
are obliged to say that he sloughs his skin like the moon. In Australia,
the natives of Queensland believe that the moon sheds its skin every
month. In Togoland, the Ewe believe that the moon has scales like a
serpent and casts them off each month; while an ancient Babylonian
inscription describes the body of the moon-goddess Ishtar as ‘covered
with scales like a snake’s’. Moon-gods and moon-goddesses are usually
represented as serpents.

In a more general way the moon is conceived as dying every month
and as being bom again after three days, and the power of men to sur-
vive every month, or the hope that they will survive after death, is
regarded as being derived from the moon. Thus the eternal moon is not
only the cause of death, but also the source of renewed life and resurrec-

THE MOTHERS

310

tion. The natives of the Caroline Islands, for instance, think that men
formerly went to sleep as the moon waned, slept during the three days
of the interlunary period, and rose again with the new moon. Accordingly
they rejoice at the appearance of the new moon, and say that they are
born again. The Sakai of the Malay Peninsula relate that the moon,
which falls to the earth during the three days of the interlunary period,
has to be assisted back into heaven by the use of magic rites. Unless this
is done, all men must die. In India it has been recognized since Vedic
times that ‘the moon makes life long’. At the full moon, in the month of
Kuar, the Hindus place food on the house-tops; when it has thoroughly
absorbed the rays of the moon the food is distributed among the family.
‘This is supposed to lengthen life. , In the Upanishads the moon is
represented as the cause and controller of metempsychosis. The Indians
of California, as they dance before the moon sing: ‘As the moon dieth
and cometh to life again, so we also, having to die, will live again.’ The
same belief obtains among other American Indians, as it did in ancient
Mexico and Peru. The ancient Babylonians prayed in much the same
manner. ‘May the gods’, they said, ‘give me a life which, like the moon,
is renewed every month.’ In Loango the women prayed to the new moon :
‘So may I renew my life as thou art renewed.’ The Christian Abyssinians
address long prayers to the new moon. ‘He is risen!’ the women ex-
claim.

In the African stories in which two messengers bring respectively the
message of death and the message of eternal life to mankind, the sender
of the message, in what appear to be the more primitive versions of the
tale, is the moon. The story is current in this form among the Bushmen.
Among the Tati Bushmen the moon is said to have given the message to
the tortoise. The tortoise being incredibly slow in delivering the message,
the moon repeated it to the hare, which perverted it so as to make it
mean the opposite of what the moon intended. The moon is the sender
of the message among the Hottentots and among the Nandi and Masai
of East Africa. In other versions of this widespread African myth the
sender of the message is simply the personage who is most prominent
in the mythology of the particular tribe or people; but it is probable that
these personages were originally the moon.

In Madagascar a current myth relates that the first men were given
the choice of dying like the moon or like the banana-tree; that is to say,
of dying periodically and being born again, or of dying altogether, but
propagating the species as does the banana-tree, by its root. The first
parents foolishly chose to propagate and die like the banana-tree, and
thus lost the chance of being immortal. This last version of the myth of
the origin of death differs somewhat from African versions, where, as far
as I know, no reference is made to the alternative of propagating like the

THE RESURRECTION AND THE LIFE 311

banana. It may have been brought to Madagascar from the Indonesian
region by the Hovas, for the allusion to the banana in similar myths is
frequently found there among the Mantras of the Malay Peninsula, the
Alfurs of Poso and on the island of Nias.

The reason the banana is chosen as an emblem of sexual propagation
is probably its phallic form. It is implied in the foregoing myths, as in
the narrative of Geneses, that men lost, or gave up, their hope of
immortality through their fatal subjection to women and to their sexual
passions. In a creation myth from the Gilbert Islands the creator, after
bringing the first man and the first woman into existence, expressly for-
bade them to have sexual intercourse with one another, although he
himself had intercourse with the woman. Propagation by sexual inter-
course thus appears to be regarded in primitive thought as an alternative
to eternal life. Mankind cannot have both; men must choose between
the moon and the banana.

In Polynesia also stress is laid on the contrast between the immortal
nature of the moon and human mortality. In various parts it is related
that Maui endeavoured to obtain the gift of immortality from the relent-
less goddess by force. The means by which he proposed to accomplish
this were identical with those by which young men in Australia, on the
Gold Coast of Africa, and in Guiana, are fabled to have changed their
skins, and thus, presumably, acquired a new life from serpents, namely
by passing through the bodies of the monsters. Maui similarly endea-
voured to jump into the mouth of Hina or, according to another version,
to enter her womb. This attempt to be born again is said, however, to
have ended in failure. In the Melanesian stories of the origin of death
which we have noted, the bestower of life and death is the moon. The
old woman who changed her skin and wished to bestow the gift of
immortality on her grandchild was the goddess ‘Round Head’ — that is
to say, the moon. The good Kambinana, who wished men to live for
ever, and the wicked and foolish Korvouva, who bestowed the gift on the
serpents instead, are the Melanesian gods of the full moon and of the
new moon.

Similar beliefs are general among the Australian aborigines. Thus the
tribes of Central Australia believe that formerly men did not die per-
manently, but rose again after the third day. The belief that black-
fellows rise up again as ‘white men’ is universal in Australia. When the
body of an Australian black begins to decay the loosened epidermis
assumes a whitish appearance; friends assist the dead man to cast off
his skin by scraping it off with shells. The ghosts of Australian black-
fellows are, in fact, white; and in the corroborees the dancers, who are
supposed to impersonate them, paint themselves white with a prepar-
ation of pipe-clay, which they call ‘the moon\ Thus in the most in-

THE MOTHERS

312

timate beliefs of peoples in the five parts of the world, the gift of re-
newed life, the hope of resurrection and life everlasting, are connected
with the moon and the serpent.

The Serpent as Representative of the Moon

Among the founders of European civilization also, immortality is de-
rived at times from the moon, at times from the serpent. In one of the
Greek myths of Glaucus, the secret of resurrection is acquired from a
serpent; in another version from the moon-hare. Endymion derived his
eternally renewed youth directly from the moon. The serpent and the
moon are similarly interchangeable in the beliefs and myths of the peoples
of Melansia, Australia, Indonesia and Asia, Africa and America. The
association between the moon and the serpent is thus clearly founded
upon the possession by both of the gift of immortality through perpetual
renewal, and the serpent is accordingly regarded as a form of personi-
fication of the moon. Amongst the Greeks it was a popular belief —
Aristotle gravely reported it as a fact — that serpents have as many ribs
as there are days in the lunar month. The Pawnee Indians believe ser-
pents to be subject to the moon. The Iroquois thought that the moon fed
on serpents. Among the Algonkin tribes the serpent ‘crowned with the
lunar crescent was a constant symbol of life in their picture-writing/
Their mythical ‘grandmother’, the moon-goddess Aataentsic, was rep-
resented either as an old woman or as a serpent. In Mexico the serpent-
woman was the moon-goddess. In Australia, serpents are the ‘dogs* of
the moon. In Uganda the sacred serpents’ festival took place on the day
of the new moon. Among the ancient Hindus the Naga kings, or Serpent-
people, who were supposed to be descended from a serpent, were known
as the Lunar dynasty.

So intimate is this association that it may safely be laid down that,
wherever we find the serpent in symbolism or worship, we may con-
fidently expect to find a lunar cult. In the religions of civilized societies,
as well as in those of uncultured peoples, serpent deities are lunar deities.
In India the great serpent ‘that had been worshipped there since the
world began’ is specially associated with the moon-worship of the ancient
Aryas. Among all Semitic peoples the serpent is the avatar of the moon
deity. The moon goddesses of Hellenic cults were associated with ser-
pents. The Arcadian Artemis, Hekate, Persephone, hold serpents in their
hands. The Erinys, Gorgons, Graia are serpent-goddesses whose hairs
are serpents. There is a popular superstition in Central Europe that a
hair plucked from a woman who is under the influence of the moon — i.e.
who is menstruating — will, if buried, turn into a serpent. The hairs of a
witch, according to Breton tales, turn into serpents. The notion is cur-
rent in Japan. ‘The myth of Medusa has many counterparts in Japanese

THE RESURRECTION AND THE LIFE 313

folklore, the subject of such tales being always some wonderfully beauti-
ful girl whose hair turns to snakes only at night, and who is discovered at
last to be a dragon’s daughter. But in ancient times it was believed that
the hair of any young woman might change to serpents — for instance,
under the influence of repressed jealousy,’ All women, being more or less
witches, are thus thought to have something of the serpent in them. It is
said in the Congo that at the time of the great flood all human beings,
through fright or otherwise, resumed their original shape; the men
turned into monkeys and the women into lizards. A mediaeval legend
asserts that women were made out of the legs of the serpent, which lost
its limbs as it entered Paradise. It is supposed to be not uncommon for
women to give birth to serpents and other reptiles. ‘Neither is it hard’,
says Dalla Porta, ‘to generate toads of women’s putrefied flowers ; for
women do breed this kind of cattel, together with their children ... as
frogs, toads, lyzards, and such like; and the women of Salerium, in times
past, were wont to use the juice of parsley and leeks, at the beginning of
their conception, and especially at the time of their quickening, thereby
to destroy this kind of vermin with them. A certain woman lately
married, being in all men’s judgment great with child, brought forth
instead of a child, four creatures like frogs, and after had perfect health.
But this was a kind of Moon-calf.’

In a Tartar poem the hero finds it impossible to kill a witch, even
though her bowels be torn out of her body, for she keeps her soul in a
snake. In China serpents are regarded as the source of all magical powers.
In Hebrew and in Arabic the terms for magic are derived from the words
meaning serpent. According to Philostratus, the Arabs held that a man
acquired magical powers by eating the heart or the liver of a serpent. In
Brittany supernatural powers may be acquired by drinking a broth pre-
pared from serpents. The Iroquois regarded serpents as the usual dis-
guise of witches, and believed that all magical powers were derived from
serpents, although their source was at the same time the moon. Among
the Algonkin tribes a ‘manitu’ was defined as ‘he who walketh with a
serpent’; and among the Siouan tribes the words ‘manitu,’ ‘wakan’
denoted both wizards and serpents. Among the Missouri Indians,
Captain Bossu saw ‘an old woman who passed for a magician; she wore
round her naked body a living rattlesnake’, to whom she spoke. Many
other examples might be cited. We have seen that in some West African
religious associations the women who take part in the magic rites must be
provided with serpents. It is not improbable that the countless images
of goddesses in Western Asia, in Krete and in Greece who hold serpents
in their hands reproduce the attitude of their priestesses, who were wont
to handle sacred serpents, from which they derived their magic power.
There can be little doubt that the ‘wisdom of the serpent’, the wisest of

THE MOTHERS

all beasts of the field, refers to the proficiency of the reptile in the arts
of magic.

Women and Serpents

As some of the foregoing beliefs indicate, women are as closely as-
sociated with serpents as they arc with the moon. In southern Italy it is
a current saying that serpents make love to all women — a belief familiar
from the most ancient times, for the ‘fauns’ of primitive Italy were
worshipped in the form of serpents ; and, as is well known, women were
in constant danger of being assaulted by them. Even the Good Goddess
was ravished by the god Faun us in the form of a serpent. The notion
was well known to the ancient Greeks. Euripides describes the women of
Thebes as washing at sacred springs, whence issue serpents that girdle
round them, fondling them and licking their cheeks. The predilection of
serpents for women was associated with the sacred functions of women
in the agricultural rites which came to be associated with Dionysos and
Orpheus. Plutarch tells us that Olympias, the mother of Alexander the
Great, who was an enthusiastic devotee of the women’s religion, ‘was
wont in the dances proper to these ceremonies to have great tame ser-
pents about her, which sometimes creeping out of the ivy in
the mystic fans, sometimes winding themselves about the sacred spears
and the women’s chaplets, made a spectacle which men could not look
upon without terror’. The ancient Celts also believed that serpents arc
attracted by women. In several stories serpents are described as fastening
upon women in such a manner that it was impossible to separate them.
Tenau ‘of the golden breast’ was so called because a serpent had clung
to her nipple so tenaciously that the breast had to be cut off' and re-
placed by one made of gold. In Germany women are said to become
pregnant by a serpent entering their mouth while they are asleep. In
northern France and Portugal at the present day women are believed to
be in constant danger of being assaulted by lizards.

In Japan and India, as in Europe, Abyssinia and Algeria, girls are
warned to be careful in approaching serpents lest they should be ra-
vished. A love-sick serpent will, it is said, constantly follow a woman.
Woman are supposed to conceive a passion for serpents, and to take
pleasure in allowing the reptiles to bite them, for they are thought to be
immune to their poison. The condition is known as ‘nar-ashakh* or
‘serpent love’. It is to serpents that women turn when desirous of off-
spring. The Brahmans ‘think that children can be obtained by wor-
shipping a cobra’. (Children who have been thus obtained are usually
named after the serpent.) It is a very ancient notion in India and in
neighbouring countries of the East that women at the time of men-
struation, or at least of puberty, before thay have menstruated or had

THE RESURRECTION AND THE LIFE 315

intercourse with a man, are possessed by a malignant spirit in the form
of a serpent.

Among the Jews it was a common Rabbinical opinion that menstrua-
tion owes its origin to the serpent having had sexual intercourse with
Eve in the Garden of Eden. The ancient Persians likewise believed that
menstruation was originally caused by the serpent-god Ahriman’s re-
lations with the first woman. The ancient Eygptians had fables about wo-
men being pursued by serpents. Among the Hottentots snakes are sup-
posed to have connection with women while they are asleep.

In several of the notions just noted serpents are regarded as being the
cause of menstruation; they thus play the same part in regard to the
functions of women as the moon. In New Guinea the idea also appears to
be familiar. Although, as we have seen, the Papuans commonly ascribe
menstruation to the moon having intercourse with girls and women, in
their elaborate wood-carvings women are often represented as being
bitten in their genital organs by lizard -shaped animals. Throughout
Polynesia women are supposed to be in danger of being violated by
lizards and by eels.

This notion has been thought to derive from the phallic shape of the
animal, and that idea is undoubtedly present in those world-wide be-
liefs. It was thought by the ancients, and is still believed by the Euro-
pean peasantry, that during sexual conjunction the male serpent in-
troduces its head into the mouth of the female, and that the latter gnaws
and bites it off', thus becoming fecundated. The same idea appears to
obtain in Polynesia, and is, no doubt, general among uncultured peoples.
In parts of the world where there are no serpents, lizards and other ani-
mals which are understood to change their skins, appear in the same
role. In most of these myths, the animals in question are regarded as
impersonations of a moon-god.

A variant of the same ideas represents serpents as stealing women’s
milk, e.g. among the Namaquas of South Africa. The same belief is cur-
rent in Madagascar and New Guinea, and has also been popular in
Wales from time immemorial. In Italy the restlessness of babies is
frequently attributed to a serpent sucking the mother’s breast, to the
exclusion of the child.

The Eskimo have stories of reptiles falling in love with women and of
serpents caressing women clinging to their breasts. The North American
Indians likewise have numerous stories of serpents having connection
with women and falling in love with them. Among the Dene, as among
the Jews and the Persians, tradition relates that the first woman mated
with a serpent.

In South America serpents are generally regarded as the cause of
menstruation. Pregnancy as well as menstruation may, of course, result

THE MOTHERS

316

from the assaults of serpents. The Arawaks believe that such accidents
are quite common. In the Upper Amazon region the god Jurupari, who
is a serpent, is said to pursue women and to ravish them; they there-
after give birth to serpents. He himself was conceived by a virgin, who
had no sexual parts at all, but was bitten by a reptile while she was
bathing in a pool. The superstitions of the Tupis of northern Brazil
bear a strong likeness to the rites of the Theban women described by
Euripides. ‘The Devil’, says Father Yves d’Evreux, ‘has persuaded
those Gentiles of various delusions concerning waters, fountains and
streams. Some are inhabited by nymphs, others by goddesses; some
produce one effect, some another; some are injurious and dangerous,
others agreeable and safe; some are sacred, others profane. Those
savages have likewise a superstitious opinion that when they see a cer-
tain kind of lizard, which resemble those we call “mourons”, or venomous
snakes, running in the waters, they esteem that the fountain is dangerous
to women, and that Giropari (their chief spirit) drinks of that water.
This superstition goes so far that they believe that those lizards cast
themselves upon women, that they send them to sleep and have com-
pany with them, so that they become big with child in consequence, and
give birth to lizards instead of children.’

Serpents the Guardians of the Waters of Life

The association of serpents with sacred springs, just noted, is common
in many other parts of the world. In an Indian tale a beautiful girl is
taken by the King of the Serpents as his wife while she is admiring her
reflection in a spring. The Zulus of Natal have a similar story. In Brazil
every spring and lake is regarded as being under the control of a serpent-
deity. So also among the North American Indians.

In innumerable myths, of which that of Perseus and Andromeda is
the type, the sun-hero liberates a woman who has been given as wife
to the serpent of the sea, or of a river, or spring. There can be little
doubt that the opponent of the sun-god in these myths was originally
the moon-god. In numerous tales the latter is credited with assaulting
or kidnapping women while they are drawing water at a spring or well,
just as the serpent-guardians of the springs. Wells and springs are, in
fact, particularly dangerous to women. In several Irish tales the water of
a spring or well overflows and pursues women. Wells also, especially very
sacred ones, are credited with the property of causing women to fall
pregnant. We have come upon the belief in South America and in
Polynesia. Kaffir women, when their children ask them where babies
come from, tell them that they are found by waters and springs, and that
‘women bring them back with them when they return from fetching the
day’s water’. Banyoro women refuse to cross rivers unprotected, for

THE RESURRECTION AND THE LIFE

317

fear of assault by the serpents dwelling in them. Recently a well near
Oxford was regarded as being almost as potent as the sacred wells of the
Tupis. ‘Childs Well,’ wc are told, ‘by the holiness of the chapleynes
successively serving there, had vertue to make women who were barren
to bring forth children.”

Whether it be the serpents which dwell in the wells and springs, or
the waters themselves, which get women with child, the original cause
of the mischief is the moon. It may appear strange that serpents should
be so generally and intimately associated with water. Serpents are not
usually aquatic animals. But primitive zoology does not look closely into
classificatory affinities : the serpent is interchangeable in fable with al-
most any reptile, and with the eel and the crocodile, and is generally
supposed to have once possessed legs ; it is likewise regarded as identical
with the worm, and is often so spoken of. Fishes are commonly regarded
as closely allied to serpents. The Bechuana, for instance, ‘have a prejudice
against eating fish, and allege a disgust to eating anything like a serpent.
This,’ remarks Livingstone, ‘may arise from the remnants of serpent-
worship floating in their minds.’ The Caribs of the West Indies appear
to have had the same notion. The Tasmanians did not eat any fish that
has scales, nor did the ancient Egyptians.

Babylonian gods were conceived in the form of serpents or of fishes,
interchangeably. Ea, or Oannes, who was the serpent Tiamat, identified
with the primal waters of life, was also the great Fish. The Syrian form
of the Semitic goddess, Derketo, was a fish, and her dying and resur-
recting son, Tammuz or Adonis, was also a fish-god, and was spoken of
as Dagon, the Fish. Similarly the Hebrews regarded Leviathan, — that is,
the primal serpent — as a huge fish covered with scales. Joshua, the first
saviour of Israel, was called ‘the son of Nun’, that is to say, ‘the son of
the Fish’. Nun, or Ji-nun, the Fish, was, according to learned Rabbis,
the true name of the Messiah, which he bore before the sun was created.
He was currently regarded as identical with Leviathan, who having
died, rose again after three days. The symbolism passed over into early
Christian ideas, and a fish, an image of which Clement of Alexandria
recommends all Christians to wear, was as common an emblem of the
faith as the cross.

The serpent’s constant association with waters is not, however, due
to its fancied resemblance to a fish, but to its being a surrogate of the
moon, the ruler of waters. Throughout Indonesia the ocean is regarded
as a serpent, the Great Serpent of the primordial waters, from which all
things have arisen. The great serpent or dragon, so prominent in Chinese
mythical representations, is the representative of the waters above the
earth. The association of the serpent with rain has led many to suppose
that this is the primary significance of the serpent in universal mytho-

THE MOTHERS

318

logy. But rain is only one of the forms of the waters over which the moon
exercises her control. The moon’s faculty of eternal renewal is fre-
quently ascribed to her bathing periodically in the waters of pools, lakes
or of the sea. In Polynesia the moon preserves her youth by bathing in
the waters of Tane. The Jukon of northern Nigeria say that the moon
goes to renew its youth in its home in the waters. So do the Huitotos
of the Upper Amazon. The water which by that baptism imparts re-
newed youth to her is thus the Water of Life. It seems not unlikely that
the virtue of those waters is thought to be due to the presence in them
of their guardian serpents, and that the moon is regarded as deriving her
powers of rejuvenescence from the serpent, which renews its skin.

CHAPTER 21

Primitive Cosmic Religion

As we have seen, primitive religious ideas are hopelessly parochial.
They are not general theories of the universe, but arc connected with
primitive man’s immediate needs and purposes. Superstitions concern-
ing the moon, however, are of especial interest since they constitute the
germ of a cosmic religion. Like the totem, the moon is regarded as the
progenitor of the tribe, and the tribe is equated with ‘mankind’. But the
moon naturally tends to broaden out into a universal deity, whose
sphere of influence is the whole world ; such a conception may be called
cosmic, in distinction from purely tribal and local ones. In this chapter,
therefore, w r e shall survey the religious conceptions of the less advanced
peoples, and shall sec how far the moon occupies the chief place in them.

There is little evidence to support the idea, once popular, that one of
the earliest forms of religious belief was ‘nature worship’. Primitives, it
is true, readily regard natural objects as the abode of spirits, but cos-
mological views do not seem to have developed out of such haphazard
‘animism’. Nor are the gods of winds and storms true deities; they are
usually simply aspects of more generalized powers. The facts suggest
that religious ideas arose from the desire to acquire magical powers and
not from an intellectual desire to interpret nature. The interpretations
subsumed under the term ‘nature worship’ appear to be late phenomena.

All religious conceptions tend to detach themselves from the ideas
from which they arise. Once a deity emerges with a name and personality,
these constitute a sufficient account of him, and the ideas from which
the deity arose are usually repudiated or regarded as a mere allegory.
The deity is, perhaps, remembered as the Light of the World, and the
sun is ‘naturally’ likened to the god. Clans which have become pre-
eminent frequently assimilate themselves to the sun, and favour the
substitution of the more brilliant luminary. The respective sexes of sun
and moon become changed. In solar theologies, the moon is belittled
and discredited; a regular controversy develops between the solar and
lunar conception of a heavenly ruler. But the attributes of the solar god
are, in point of fact, just those which have been handed down to him
from the older lunar deity. In primitive thought the sun is destitute of
significance. Equinoctial time, and the relation between the seasonal

THE MOTHERS

320

changes and the course of the sun, are unknown to primitive humanity.
The most uncultured races do not realize that daylight is due to the sun;
it is believed to be an independent phenomenon. The sun’s brilliancy
and heat — the attributes which constitute the manifest grounds for its
importance — count for nothing in primitive ideas of the supernatural
which are not a result of the impressiveness of natural phenomena but
an endeavour to secure benefits by obtaining or controlling magical
powers. It is the moon, the regulator and transformer, and not the sun,
which is everywhere regarded as the source of magical powers ; no primi-
tive magician turns to the sun to obtain that power. In a Lettish myth
the moon is represented as saying to the sun: ‘What have you to be
proud about ? Is it that you give heat to the world ? I do more, I cause
all growth and all existence, I bring to life and I bring to death.’ The
solar gods of later stages of development are but transformed moon-
deities, whose functions they inherit. They are life-giving and fertilizing,
they renew their youth in cycles. But such functions are the attributes
of the primitive lunar deity. Hence it is that we repeatedly come upon
incongruities which disclose the substitution: we encounter sun-gods
who die a monthly death and rise again after three days, or sun-cults
which are celebrated by night.

Nor are solar cults alone derived from older lunar cults. The more
highly developed religious systems of advanced cultures, which have
the universal bearing and philosophic scope which we associate with a
religious conception, also betray their development from lunar cults . 1

Cosmic Religion in Melanesia and New Guinea

The Melanesian islands provide an unequalled opportunity of ap-
proaching the root of primitive cosmic religious conceptions — although
to be sure Melanesian religious ideas certainly are not primitive in the
absolute sense. The Melanesians have a copious oral literature of myths,
all of which, from the northernmost groups of New Britain and the
Admiralty Islands to New Caledonia and Fiji, repeat essentially the
same themes. These themes are exclusively lunar ; the sun only appears in
a quite subordinate character.

1 ‘Observers’, remarks Sir James Frazer, ‘ignorant of savage superstition,
have commonly misinterpreted such customs as worship or adoration paid to
the moon. In point of fact, the ceremonies of new moon are probably in many
cases rather magical than religious.’ True, the rituals by which primitive human-
ity endeavours to utilize and acquire the virtues which it believes to reside in the
moon may be regarded as operations of magic. Yet if procedures which are
intended to secure the continuance of life, the redemption of human nature by a
new birth, and which are addressed to a power which is regarded as the con-
troller of human destiny, and as the resurrection and the life, are not religion, it
is difficult to say what is.

PRIMITIVE COSMIC RELIGION

321

The lunar deities of Melanesia are characteristically represented as a
pair of gods, corresponding respectively to the lucky waxing moon and
to the unlucky waning moon. The one is beneficent and wise, the other
maleficent and foolish. Thus, at Aurora, in the New Hebrides, ‘Tagaro
wanted everything to be good, and would have no pain or suffering;
Suqe-matua would have all things bad. Whatever Tagaro did was made
right, Suqe was always wrong ; he would have men die only for five days.’
Whatever Tagaro ate increased as he ate it, but Suqe could not plant
things aright. Suqe is the ruler of the dead. His head is forked. In New
Britain the two brothers are called Kabibabam, ‘the Wise,’ and Karvuvu,
the ‘Left-hand god\ (In the southern hemisphere the new moon cres-
cent appears on the left-hand side of the moon to an observer facing
north, the dark shadow being on the right-hand side.)

The lunar character of these deities is shown in many small ways;
sometimes they are known as skin-changers, sometimes they are glo-
bular in shape and sometimes the deity has eleven brothers. They create
men and constantly resurrect them after three days. Often the bad deity
seduces the wife of the good deity.

The bright moon-god is commonly regarded in Melanesian myths as
corresponding to the full moon. In the island of Erub in the Torres
Straits, the moon is represented by two sacred stones, one circular, the
other crescent-shaped. The same symbols recur in almost every re-
ligion. In ancient Egypt, almost every high god and goddess bore the
emblems of the full orb and the crescent ; these two symbols have often
been supposed to represent the sun and the moon — but we find the
emblems surmounting the figure of the moon-god. Similarly, the various
attributes of the waxing, or full-moon god, as we find them in Melanesia,
may easily be adapted to a solar mythology ; all that is needed is to inter-
pret the brilliant and beneficent full-moon deity as the sun, and to leave
unchanged the contrast between him and the deformed, dark, maleficent,
crescentic moon-god. Thus, according to Mrs Hadfield, the natives of
the Loyalty Islands worshipped the sun and the moon. ‘They say the sun
reached old age in one month, but owing to some secret and special
attribute was enabled to renew his youth.’ On the notion of a solar deity
who passes through a monthly cycle of ageing and rejuvenation, com-
ment is superfluous. Reporting from Pentecost Island the myth of
Tortali, who absents himself from the heavens for several days at inter-
vals while with his wife. Father Suas refers to Tortali as the sun, but it
is difficult to see on what principles of nature symbolism the sun is
represented as being absent from the heavens during the interlunar days,
which the moon is usually supposed to spend in the company of his wife,
the sun.

The two contrasted moon-gods of Melanesia are often associated with

t.m. — 11

THE MOTHERS

322

a moon-goddess, their mother, who is sometimes represented as giving
birth to them by splitting into two. In New Ireland she is spoken of
simply as £ the Mother’, or ‘our grandmother’. In New Britain she is the
mother of Kaninana and Karvuvu, whom she moulded out of her blood,
and is known as ‘the Old Woman’, or ‘the Shining Woman’. In the Banks
Islands she is the mother of Qat and the Tangaros, and is called ‘Round
Head,’ or ‘the Skin-changing Woman.’ In San Cristoval she is the Snake-
woman and the mother of mankind. But the serpent deity of San
Cristoval, though regarded in its local forms as a female, is in its generic
form a male, and has a brother ; it is thus a form of the double moon-god.
It is regarded as the supreme deity, the creater of the world and of all
creatures, and is the object of deep reverence. The Great Serpent, the
creator of mankind and of all good things, occupies the same position in
the Admiralty Islands, and in Pine Island, at the opposite extremity of
Melanesia. The same triad of the two moon-gods with their mother is
also found in New Guinea.

A misconception which has had a paralysing influence on the inter-
pretation of primitive belief is the assumption that a given personage
cannot be a god because he is not treated with reverence, or because
ludicrous stories are told about him. Such beings, it is supposed, must
at best be classed as ‘demi-gods’, or ‘heroes’. But these distinctions, so
important to the Christian theologian, do not exist for the savage. He
makes no clear distinction between gods and men. The magical attri-
butes of supernatural beings are but a superlative form of the powers
daily exercised by medicine-men ; the function of creator does not estab-
lish a gulf between the god and the man; all magicians are more or less
creators, and the circumstance that he has created the world is often an
incidental detail, hardly worth stressing. As often as not, the drudgery of
creating beings and things is delegated by the god to a subordinate
demiurge. The god is commonly supposed to have dwelt at one time
amongst men, living much the same life as his savage companions, and
changing his residence to heaven by a mere accident in the course of his
adventures. The character of awful majesty, which we associate with a
veritable god, the attitude of reverence which we regard as characteristic
of a religious view of him, are foreign to primitive man’s conception of
supernatural beings. These may be dreaded as dangerous and evilly
disposed, but dread does not prevent a sly delight in their misadventures ;
they may be looked upon as well-disposed and willing to be helpful, and
are then treated all the more freely with friendly good-nature, but not
with reverence (cf. ‘the man in the moon.’) Thus it is that the gods of
Melanesia and of Polynesia have commonly been described as ‘tribal
heroes’ or ‘spirits’.

Melanesian secret societies, when they first came under the notice of

PRIMITIVE COSMIC RELIGION 323

Europeans, were already in an advanced state of decay, but enough of
their primitive character survives to show that their central object was
admission to the company of deceased ancestors by obtaining resur-
rection and eternal life from the moon. ‘It is possible 5 , observed Dr
Rivers, ‘that further knowledge will show the presence of features de-
rived from the cult of the moon in the ritual of Melanesian “ghost”
societies . 5 And, in fact it does. At Vanikoro, in the Santa Cruz group,
where, as everywhere else in Melanesia, the creation of the world and of
the human race is ascribed to the moon; its cult in the stone circles con-
stitutes, in fact, a regular religious worship.

Thus the Qat, Qatu or Qetu societies of the Banks Islands and north-
ern New Hebrides are associations of the moon-god Qat.

In Rook Island, between New Britain and New Guinea, the boys
were supposed to be swallowed by a monstrous spirit, and were de-
livered shrieking and trembling to the masked personages representing
it, and compelled to crawl between their legs. In the Koko Islands the
candidates for initiation are seized, blindfolded and thrust between the
jaws of a vast crocodile effigy, while their parents shout: ‘Do not kill my
child ! 5 Bull-roarers swung inside the building represent the roaring of
the monster. At last, yielding to the entreaties of the agonized parents,
the master of ceremonies consents to beg the crocodile to disgorge his
prey, on consideration of a payment of some pigs as a substitute for the
boy, who is accordingly vomited forth a fully-initiated man. Almost
identically similar proceedings constitute the initiating rites of the
Yabim, the Bukua and the Kai tribes of northern Papua.

It would be a misconception to imagine that these proceedings are
mere pantomines designed to instil terror into the uninitiated. The mon-
ster is the most sacred and venerated object in the religious ideas of
those tribes. Offerings and sacrifices are presented on solemn occasions
to the beasts; and they are consulted as oracles. After a war expedition,
prisoners are immolated and their bodies are placed in the stomachs of
the wicker-work crocodiles; they are subsequently removed and eaten
communally by the men. The holy crocodiles are called ‘kopiravi 5 , and
are spoken of as ‘kai ai imunu 5 . that is to say, ‘imunu from the sky 5 . The
term ‘imunu 5 is explained as meaning ‘the principle of life 5 .

The crocodile monster which devours and regurgitates the candi-
dates for initiation is thus no other than the ‘kopiravi 5 , or ‘principle of
life from the sky 5 . In the Kiwai tribe, the initiators representing the
‘spirit 5 are masked as crocodiles. This conception, ‘which would appear
to correspond to the “mana 55 of the Melanesian Islands , 5 says Professor
Haddon, ‘runs through all their religions . 5 The crocodile occupies, in
fact, a conspicuous place in Melanesian New Guinea. In the Solomon
Islands, says Dr Codrington, the sacred animals ‘are chiefly sharks.

THE MOTHERS

324

alligators, snakes, bonitos and frigate-birds’. The shark, and the ‘bonito’
fish which resembles it, are in all likelihood regarded as marine equiva-
lents of the crocodile. In Polynesian languages the shark and the lizard,
the nearest Polynesian equivalent to the crocodile, are both denoted by
almost the same word, ‘mako’ and moko’, respectively. (We have seen
that women are represented as being bitten in their genital organs by the
crocodile.)

In New Guinea initiation ceremonies the candidates are often said to
be restored by the monster which has swallowed them — on considera-
tion of some pigs being supplied by the parents as a ransom. This is not
exploitation on the part of the hierophants — though, of course, it may
degenerate into such. In some tribes, in which the ceremonies have
become reduced to a simplified form, the final investitute of the can-
didate takes place while he stands on the carcass of the dead pig which
he has provided, and, before this, he is steamed as if he, and not the pig,
were about to be eaten. The slaughtered animal is thus regarded not
merely as a payment, but as a substitute , whose life is given in ransom for
that of the candidate. The Namau offer their prisoners of war to the
‘kopiravi’ crocodiles in the same manner as the ancient Celts. In the
sacred stone circles at Wagawaga and Wani a man was solemnly de-
corated, tied to a stone, sacrificed, roasted and ritually eaten. In these
forms of the ritual the substituted victim is frequently assimilated to the
deity itself; the Papuans doubtless considered that they acquired the
virtues of the deity by eating the sacrificial victim. In the Solomon
Islands it is said that formerly men cut up the moon and each ate a
piece of it. The substitution of a redeemer who gives his life for the
people is not infrequently enacted symbolically in the same manner as
the symbolic death of the candidate for initiation is represented by
a mimic death.

In Melanesia the lunar power is associated with sacred stones. ‘Sacred
places have almost always sacred stones in them,’ and deities are associ-
ated with them. As already noted, in the island of Erub, the moon is
represented by two sacred stones. In the Solomon Islands, the creator,
Tantanu, is regarded as a stone. In Fiji the chief god, Ndeugei, is a serpent
in the upper part of his body and a stone in the lower. The goddess
‘Round Head’, the mother of Qat, is represented in the Banks Islands
by a stone. On Lepers’ Island ‘all the stones that are sacred are con-
nected with Tagaro’. On San Cristoval a magic stone spoke and an-
nounced the coming of the Great Serpent, and was in fact his daughter. In
the Massim area of eastern New Guinea, the circles of standing stones,
which were of old the scenes of human sacrifice and cannibalism, are
marked with the remains of ancient pictographs; one bears a cross;
another is called ‘the Serpent*. Dr Seligman was unable to imagine why,

PRIMITIVE COSMIC RELIGION 325

as it is wholly unlike a serpent. In New Caledonia all magic and re-
ligious practices are associated with sacred stones, and these bestow
fertility. In the Belep tribe the sacred stone is regarded as ‘the principle
of life’. The god Doibet, who is ruler of the dead and of the underworld
and is manifestly the representative of the dark moon, is represented as
being, in the lower part of his body, a stone. These sacred stones play
the same conspicuous part in every district of Melanesia, from the Ad-
miralty Islands to Fiji. Moreover, individual fetiches are adopted by
selecting a stone from which the owner believes he derives magic power,
or ‘mana’.

Sacred stones are a world-wide feature of archaic religious cult, from
Japan to Brazil and from Madagascar to Scandinavia. Every Semitic
deity was embodied in a sacred stone, and could be called down by the
worshipper by setting up a stone; aniconic stones played a conspicuous
part in primitive Greece. Sacred stones represented the gods in all Celtic
countries ; and they are found throughout Africa.

It might seem that stones afford such an obvious means of setting up
a durable symbol that the practice of ‘worshipping’ stones scarcely calls
for any explanation. Yet it appears that the identification of stones with
the moon comprises something more. The sacred stones of Melanesia
are not the aniconic idols of a people unable to produce more like-life
representations of their gods, for few savages excel the Melanesians in
wood-carving. In the island of Yam, as already noted, two deities,
doubtless forms of the double moon-god, are represented by elaborate
images of a shark and a crocodile adorned with tortoise-shell and bird-
of-paradise plumage and painted in brilliant colours. But the richly
carved figures are merely presentments; the actual souls or spirits of the
divinities are situated in two rough stones which lie beneath the images.
In the myth of the Alfurs of Celebes the symbol of immortality, instead
of being the moon or the serpent, is a stone ; and we are told, in the name
of the moon-god, that the stone is the symbol of immortality because,
like the moon, it is not subject to decay. In New British myths, mankind
was defrauded of the gifts of immortality, not only by the serpents but
also by stones. The Melanesians do not merely represent their gods by
stones, but as being stones. In short, it appears probable that they are
deliberately identifying the ‘principle of life’ with stones because, like
the serpent, stones partake of the moon’s gift of immortality.

Australia and Tasmania

Australian religious conceptions do not centre exclusively round the
totem. The control of life and death and the resurrection of the dead are,
throughout Australia, regarded as depending upon the moon. The
Queensland tribes and the Dieri regard the moon as the creator of

THE MOTHERS

326

mankind. The sun has no place in Australian conceptions of the super-
natural; when personified it is regarded as a female, presumably the
wife of the moon.

The initiation rites of the Australians have both practical and
magical purposes. I have suggested that the practical purpose may be to
aid in sexual selection. Be this as it may, these ceremonies, like all rites
of initiation, whether into ‘secret societies’ or into bridal manhood, are
regarded as conferring certain magical powers, and admission to the
ranks not only of living tribesmen but of their departed ancestors, after
the initiate’s death. The latter object is achieved, as usual, by ceremonies
in which the initiate is supposed to die and be resurrected.

The method of resurrection often resembles the Australian story al-
ready noted, in which a youth is supposed to acquire a new skin through
being swallowed by a serpent, and those initiation rites of Indonesia and
New Guinea in which the neophyte is imbued with new life by passing
through the body of a monster. The supernatural being is supposed to
swallow the candidate or to bite his head off. The performance is some-
times represented in a fairly realistic manner. ‘I have seen one of the
old men’, says Dr Howitt, ‘rush furiously at one of the novices, seize
him by the head and apparently bite part of it.’ Mathews says that ‘be-
fore cannibalism ceased to be practised by the tribes he studied, ‘it was
the custom to kill and eat a man during the “burbling” ceremonies’.

The Australian ‘bora’ divinities, in whom some authorities have
thought to detect moral and august deities 1 and who thus initiate the
youth are, in fact, not remarkable for either their moral or august charac-
ter. In these myths, they break almost every one of the ten command-
ments. Their connection with the moon is evident. Baiame ‘appears
occasionally during the day, but mostly by night’; and is described as
painted white with the pipe-clay which is known as ‘moon’. On the
bora ground he is pictured by a huge figure moulded in mud and sand ;
at its head stands a tree on which is represented the lunar crescent.
Moreover, the magical powers of Australian medicine-men are inti-
mately associated with the possession of pieces of rock-crystal, which are
supposed to be obtained directly from the heavenly magicians. They are
symbols of the Great Spirit. Daramulum is represented with quartz
crystals on his head, and Baiame with similar crystals growing from his
shoulders.

1 Westermarck cites, as the piice de resistance of his account of Australian gods,
a paper by J. Manning from the Journal and Proc. of the Royal Society of N.S.
Wales (1882, p. 157 sqq.). Actually ‘Cockatoo’ Manning was a farmer who
obtained entry to a meeting of the society and read a paper which was a farrago
of nonsense. Under the rules of the society, the paper had to be printed despite
the protests of members of the society, who had heard it with mixed amusement
and disgust. Its worst puerilities were deleted before publication.

PRIMITIVE COSMIC RELIGION

327

Seeing that various Australian tribes are stated to regard the moon as
their creator and father, and that the origin of death and the faculty of
resurrection are even more generally ascribed to the moon, it seems no
very daring conjecture that their ‘Great Spirits’, the source of all magical
powers, who are seen at night adorned with moon-paint and crystals,
who are represented with crescents over their heads, and who kill the
young men in order to resurrect them, are none other than lunar deities.
They appear to be, in fact, essentially identical with the moon gods of
Melanesia and Papua. The two best-known Australian bora deities,
Baiame and Daramulum, are an associated pair. Baiame occupies the
superior position, Daramulum acting under his orders as his deputy or
demiurge in the initiation ceremonies. Baiame is represented as white,
Daramulum as ‘one-legged’ and deformed; Baiame is on the whole
benevolent or at least innocuous, Daramulum is evil and maleficent.
They stand in fact in the same relation to each other as the bright full
moon and waxing moon-god and the dark waning moon-god of Mel-
anesia. Indeed, there is substantial evidence that the Australian initiation
deities and their rites are culturally connected with the correspond-
ing cults of Papua and the Torres Straits. The bull-roarer and the
supposed ‘swallowing’ of the candidates are prominent features of both.
The bull-roarer which in Papua as in Australia is the voice of the initiat-
ing ‘spirit’, is in the Borli tribe of Papua called ‘bora’. It is difficult to
avoid recognizing in Bomai and Malu the Australian Baiame and Dara-
mulum. Malu, the ‘bad’ god, is stated to be like Daramulum in the
habit of biting off men’s heads. Malu is a crocodile; Daramulum is an
iguana — there are no crocodiles in Australia. Bomai is a shark; in Queens-
land Baiame is a tortoise.

The death and resurrection which the rites are supposed to effect are
sometimes dramatically represented. One of the older men is buried in a
deep grave, which is then lightly covered with branches and earth; at
the conclusion of the ceremony he rises again, bearing in his mouth the
magic crystal which he is supposed to have received from Daramulum.
By a converse argument the candidates, instead of being swallowed, may
be made to swallow a piece of the magic rock-crystal, ‘the symbol of the
Great Spirit’.

Here a short digression on the meaning of these sacred crystals may
not be out of place. The Nutka of British Columbia also achieve rebirth
and initiation by ingesting pieces of rock-crystal ; the Butwa of the Congo
acquire immortality in the same way. The rite has analogues in the eating
of the bark of the moon-palm by South American ‘pajes’, or the drinking
of a moon-draught by the Thonga medicine-men. All stones partake of
the lunar nature since they are everlasting and indestructible, and hard,
crystalline gems which seem to emit light especially so. In fact, gems are

THE MOTHERS

328

regarded by the wise men of the East as owing their virtues to the moon ;
thus pearls are the products of the moon, emeralds of the waxing moon,
cat’s-eye stones of the waning moon. Other gems are associated in
astrological lore with various planets and stars ; but, as we have seen, the
magic virtues of these heavenly bodies are supposed to have their source
in the moon. Furthermore, it is widely believed, — e.g. in China — that
serpents carry in their head a magical gem. The notion was familiar to
the ancient Germans, and the ‘snake-stone’ is mentioned in some of the
most ancient German poems. In Salzburg and the Jura Mountains it is
believed that one of the eyes of the undying serpent is a diamond. The
ancient Greeks believed likewise. Diamonds are popularly supposed in
France to be produced by snakes, while the Danubian peasants believe
that all gems have their origin from serpents ; a serpent’s nest is sup-
posed to contain a wealth of precious stones. Among the ancient Celts
those ‘snake-stones’ or ‘serpents’ eggs’ played an important part in
religious belief, and bestowed magic powers on the possessor. The
veneration for such ‘snake-stones,’ ‘adder-stones’ or ‘clach-nathrach’ is
still prevelant in all Celtic countries.

The Melanesians of eastern New Guinea hold similar views; the in-
dispensable means of acquiring magical power is, they believe, to obtain
a snake-stone. Among the American Indians it was believed that certain
mythical serpents have horns on their heads, and possession of one of
them bestowed magical powers. The Armenians tell of a queen serpent
who paralyses her enemies by means of a magic stone which shines
brightly when, on certain nights, she takes it in the air. Another Armen-
ian tradition tells of a man who unknowingly married a serpent-woman.
An astrologer revealed to him the true nature of his consort, and advised
him to shut her up in the oven and roast her. This he did, and her ashes
were found to possess the property of transmuting metals into gold. The
magic ‘snake-stone’, or ‘moon-stone’, is in fact no other than the philo-
sopher’s stone, and naturally derives its power of transmuting baser
metals into incorruptible gold from the moon. The true formula for
transmuting metals into gold is to treat them with a preparation ob-
tained by burning a serpent to ashes ; alchemists were commonly spoken
of in the Middle Ages as ‘serpent-burners’.

The Nutka of Vancouver, the Congo natives, and the Australian
aborigines, when they endeavour to renew their life by swallowing rock-
crystals, are thus guided by the same ideas as the mediaeval alchemists.

The ceremony which constitutes the central rite of Australian boras,
and which takes the place of the immolation and resurrection of the
candidates, is the knocking out of one or two of their front teeth. This is
represented as a commutation of more severe proceedings. It is said that
Daramulum always pretended to kill the boys and resurrect them, but

PRIMITIVE COSMIC RELIGION

329

that, in reality, he contented himself with extracting a tooth. The
evulsion of teeth thus appears to be regarded as equivalent to being
swallowed and restored to life by the initiating spirit. The evulsion or
filing of teeth at initiation is a widely distributed custom.

The teeth are, according to many uncultured races, the part of the
body in which the immortal soul resides. This is in accordance with
savage logic. The most enduring or immortal part of man is regarded as
being his bones. The Choctas say that ‘the real seat of the human soul
is in the bones’. The Hurons likewise called the bones ‘the soul’. The
ancient Hebrews appear to have had similar notions ; the word for ‘bone,’
‘etzem,’ meant with them a man’s self, his person. Among the Central
Australian tribes, the word ‘kutchi’, means both ‘bones’ and ‘ghost’. The
bones of the dead are accordingly the object of elaborate attention the
world over. The bones which are regarded as possessing in the highest
degree the quality of immortality are the lower jaw and teeth. In Uganda
the jawbones of the kings are alone preserved ; they are spoken of simply
as ‘the King’, and are consulted in all important affairs. The Ewe of Togo
state that babies are formed out of the jaw-bones of the dead, thus rein-
carnating them. Warriors adorn their trumpets and drums with the
jaw-bones of their slain foes, thus effectively conciliating their ghosts.
For the same reason, in Porto Novo, the executioner keeps the jaw-
bones of his victims securely in his house.

This conception parallels the view that stones possess the virtue of
immortality; and that the teeth should be accounted as possessing that
virtue in the highest degree follows by the same logic which sees in
crystals the most immortal order of stones. The bones of the dead are
referred to in Mayan codices as ‘precious stone bones’. In Samoa the
spirit of a shark-god was supposed to be immanent in a shark’s teeth which
were consulted as oracles. The holy tooth of Buddha is similarly wor-
shipped in Ceylon. The dragon’s teeth which Kadmos sowed, from
which armed warriors arose, probably contained the spirit of the divine
animal. Teeth are also associated with the moon in primitive thought,
because they possess to a certain extent the property of being renewed.
Among the natives of the Loyalty Islands ‘it seems to have been the
general opinion that the moon possessed an unlimited supply of teeth
to dispose of; so that whenever a man pulled out a decayed tooth he was
mindful to throw it over his house on a moonlight night, calling out at
the same time to the moon : ‘Here is my old tooth, take it away and send
me another in its place’. This faculty of renewal is shared by the hair,
the growth of which is commonly believed to depend upon the moon.
Accordingly, the cropping of the hair or complete depilation is an
essential part of the Australian rites of initiation. When, therefore,
Daramulum bites off the teeth of the initiates as a compromise for biting
t.m. — II*

THE MOTHERS

330

off their heads or swallowing them bodily, he is selecting the most
appropriate substitute for the whole man, and the portion best suited
to receive the gift of immortality.

In short, among the Australian aborigines, who exhibit one of the
most primitive surviving examples of religious conceptions, rites and
conceptions of lunar origin are no less important than those associated
with totemic animals. The same appears to have been the case with the
now extinct natives of Tasmania.

Traces of Lunar Cults in Indonesia

Where agriculture has become the chief means of subsistence, the sun
naturally assumes a special importance. Since the races of the Malay
Archipelago are mostly dependent upon primitive agriculture, the most
general cosmic conception amongst them is that of Father Su, the hus-
band and fertilizer of Mother Earth. Their religious theories have, how-
ever, been influenced by Hinduism and Islam, as shown by the fact that
the Bataks, the most primitive race of Sumatra, have a trinity which in
some respects conforms in nomenclature to the Hindu triad, while even
the wild tribes of Borneo Dayaks call their supreme being Hatalla or
Magatalla, which is simply the Arabic Allah ta’ala.

Nevertheless, more ancient ideas everywhere gleam through such
advanced concepts. Thus, the triad of gods of the Toba-Bataks, al-
though adapted to Hindu ideas, is known as the ‘Lords of the Moon. 5
With them is associated a moon-goddess, who, having bestowed upon
her children the power of controlling demons, retired to the moon to
watch over them. ‘When we are in need, in trouble, or in anxiety,’ say the
Bataks, ‘when a child, a woman, or an oppressed person amongst us is
in need of help, we look to the moon, and take courage.’

This strange combination of primitive with later cosmic conceptions
is manifested in the Kei Islands and in Aru even more incongruously;
the chief god is spoken of as the ‘Sun-Moon’. In Ceram-laut, and in the
island of Gorom the moon is feminine; but the women address her as
‘Our Grandfather’.

In the sixteenth century Barthema noted that the moon was wor-
shipped in Java. In the Kei Islands and in Timor the moon is the chief
object of worship. In Timor she is served by aged vestals, who tend an
undying fire. In the island of Babar the supreme god dwells in the moon.
In Nias the cosmogony is purely lunar. In the Middle Celebes, among
the Bataks of the Tomori region, sacrifices are offered to the moon at
harvest-time, with songs in which it is addressed as ‘Mother’. Among the
Dayaks of Borneo, we are told, the ‘veneration for the moon, formed the
chief basis of their worship and myths’. When the Spaniards first
visited Borneo, the only worship which they noticed amongst the natives

PRIMITIVE COSMIC RELIGION 331

was that ‘they pray to the moon, and ask her for children, abundance of
cattle and of the fruits of the earth, and other similar things’.

The moon is regarded by the Dayaks as the cause of time and of every
event. ‘It is certain that solar mythology was originally foreign to all
purely Indonesian peoples.’

Survivals of Primitive Cosmic Religion in Polynesia

The decay of the original conceptions upon which Polynesian myths
were founded has proceeded much further than in Melanesia. Though
the Polynesians preserved their myths with wonderful unity, their in-
terpretation was not an object of religious or of magical interest. Hence,
so far as they are ‘interpreted’ at all, they are interpreted ‘naturally’:
that is, in accordance with the more obvious aspects of natural pheno-
mena rather than with their primitive religious and magical significance.
Solar interpretations, if not insisted on, are at least accepted. This has
given European inquirers the impression that Polynesian myth presents
a typical example of the primitive ‘solar religion’ imagined in the theory
of ‘nature worship’, and the Polynesians have even been sometimes
termed ‘The People of the Sun’.

Actually there is no trace of sun-worship or solar mythology anywhere
in Polynesia. The sun plays no part in Polynesian myth, except as the
wife of the moon. The most prominent mythical figure is Maui, the
‘creator of land,’ the ‘creator of man’. He has been identified with the
sun; he is, however, the son of the moon, Hina; and the most famous
of his myths relates how he conquered the sun. Maui is demonstrably
identical with the moon-gods of Melanesia. His name means ‘left-
handed’; and he is thus, like the ‘left-handed Karvuvu’ of New Britain,
the new moon-god that grows from the left (west) in opposition to the
‘right-handed Kabinana’. His name is also used as a synonym of
‘witchcraft’.

As in all traditional mythologies, the same deities are known in Poly-
nesia and in Melanesia by various names, and the traditional names be-
come transferred from one deity to another. An instance of the con-
fusion which may thus arise is presented by the myth of Rona, who in
her best-known form is the woman in the moon, abducted there by the
moon-god, while she was drawing water. In other versions, however,
Rona is a male and an ogre. ‘Rona is lord of the sun and moon. Rona
eats the moon and the moon eats Rona; but as each becomes exhausted
and devoured in the monthly battle, they go to the Life-waters of Tane
to bathe and be restored to life and strength, by which they become able
to renew the struggle.’

The two moon-gods, the favourable and the unfavourable, not in-
frequently change places. Maui-tiki, who figures in western Polynesia

THE MOTHERS

332

as a good-natured hero anxious to confer benefits upon mankind, is in
Tahiti a fiend who desires to slay men.

The wife of Maui is Rohe or Rau. ‘She was beautiful as he was ugly,
and on his wishing to exchange faces with her, she refused his request.
He, however, by means of an incantation, managed to gain his point. In
anger she left him, or refused to live any longer in the world of light, but
proceeded to the under-world.’ As in the parallel Melanesian myth, the
wife of the moon-god is the sun. In the Solomon Islands, in the Loyalty
Islands and in New Guinea, the sun and the moon formerly travelled in
company, but quarrelled and separated ; in New Zealand the moon said
to the sun, ‘Let us travel together at night’; but they disagreed and
separated. Rohe, the wife of Maui, is in fact said in New Zealand and in
the Chatham Islands to be ‘the sister of the sun’; but it appears manifest
that, like Ro Lei, she was in reality the sun itself. In Polynesian myth,
the supposed sun-god is thus, under the thinnest of disguises, a female
and the wife of the moon.

In western Polynesia, Tangaroa occupies the place which Maui
occupies in eastern Polynesia. A Tongan version describes four Tan-
garoas, and associates them unambiguously with the four phases of the
moon. Tangaroa has, of course, been claimed as a sun-god. In Samoa he
dwells in the moon. In Hawaii and in eastern Polynesia, Tangaroa is an
evil and maleficent god; he is the ruler of Hades; in the Marquesas, he
is a god of darkness and is defeated by the god of light, Atea. He is
identical with the Tangaroas and Tagaros known throughout Melanesia.
‘It would be quite erroneous’, remarks Professor Schimdt, ‘to regard
him [Tagaro] as a loan from the Polynesian Tangaroa. He is, of course,
identical with him — not, however, as an imitation, but as an earlier form
of him, which has more distinctly retained its lunar character.’

Maui, Tangaroa and other corresponding Polynesian gods are, like
those of Melanesia, incarnate in serpents, lizards and eels, and in stones.
The deities of Polynesia and their mythology are in fact identical with
those of Melanesia.

Polynesian kings were primarily ‘sacred’ kings, — that is, representa-
tives of the gods — from whom their power was derived. The Tuitonga,
the sacred king of Tonga, was regarded as the direct descendant of Maui,
Tangaroa and the goddess Hikuleo. Thus they were associated with the
moon. In Samoa, where the name of a chief might not be pronounced,
least of all when he was dead, the phrase by which his demise was an-
nounced was: ‘The moon has fallen.’

Few particulars are available concerning Maori initiation rites but,
there can be little doubt that they included a symbolic representation of
death and resurrection. In the sacred enclosures human sacrifices were
offered ; the officiating hierophant was attired in a shroud. After initiation

PRIMITIVE COSMIC RELIGION 333

the candidate, painted red, received a new name. The death and resur-
rection of the god himself formed an important part of the rites of the
Areoi, although the ritual was of a solar and not a lunar cycle. But the
god whose death and resurrection were thus celebrated — mostly, it
appears, by night — was the moon-god of Polynesia and Melanesia.

The dualism represented by the two moon-gods, the bright and the
dark, which is so conspicuous in Melanesian and Polynesian myth, and
of which we find clear traces in Australia, is met with everywhere in
primitive lunar cosmological conceptions. This is not the expression of
a sense of the opposition between good and evil forces, but derives in
every instance from the contrast between the two beings who, in
primitive belief, are supposed to contend for mastery in the course of
the lunar cycle, the one loving darkness rather than light, and seeking
to destroy the other who would bestow light and the power of regen-
eration upon the world.

Primitive Cosmic Religion in North America

This contrast is expressed by the Greenlan

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "The Mothers (1927)"

See other formats